How Kubuntu Did Not Change

There appears to be some confusion regarding the meaning of yesterday’s announcement that Kubuntu 11.10 is going to be the last release Canonical is offering commercial support for.

For those who have not yet read about it, let me quickly recap the situation. Up until now Kubuntu was a Canonical supported flavor of Ubuntu. This essentially means that you can buy a support contract from Canonical to help you with your Kubuntu infrastructure. Every once in a while Canonical would stamp ‘LTS’ on a Kubuntu release to indicate that they would support this release for 3 or 5 of years to come (delivering security and major bug fixes primarily). The 11.10 release is the last release for which Canonical offers these services. As a direct consequence Jonathan Riddell, a good friend of mine and fearless leader of Kubuntu, will work on other technology during work hours.

You might have noticed that I was writing a lot about Canonical just now, and the reason for this is that the change mostly is about Canonical and not Kubuntu.
Kubuntu is and always has been a mostly community driven project. To give you an idea what mostly means in this case: out of the 25 people who notably contributed in the past year, 1 person was employed by Canonical to do so (i.e. 4% of general Kubuntu work was financed by Canonical). Please do not get me wrong though. Jonathan is a great developer and does a considerable amount of work, particularly in those areas where the community currently lacks motivation, hence some workflow revision is in order to make the ‘new’ Kubuntu equally efficient.

So what changes for real?

  • No commercial support from Canonical for future releases.
  • Jonathan Riddell will work on non-Kubuntu stuff during work hours.
  • Alignment of Kubuntu with other siblings like Edubuntu, Lubuntu and Xubuntu.
    For those who care: on a technical level this means that a considerable amount of Kubuntu maintained software will be moved from the main to the universe archive.
  • Probably some workflow changes that are yet to be discussed.

Is this bad?
It probably is if you wanted to adopt Kubuntu in your company and were counting on a Canonical support contract. However this is probably more of Canonical’s loss than your’s. As noted earlier there is a pool of more than 25 people one could employ directly to get the same result, perhaps even better. It is certainly sad that Jonathan will not be able to continue getting payed for working on his baby though.

Is this good?
Moving to universe bares a great deal of opportunities for Kubuntu. Primarily it gives the community yet bigger control over what the distribution looks like as we do not need to get software approved to be worthy of Canonical’s support. At the same time it also reduces the policy overhead (main inclusion for those who have heared of it). The detanglement allows us to move even closer to KDE without having to worry about conflicting interests, as what is good for KDE is not necessarily what is good for Canonical.
All in all I expect Kubuntu to become more agile and continue to regularly deliver an easy to use Linux distribution featuring the latest and greatest KDE software.

There is an occasional and not very amusing urban myth that Kubuntu is a stepchild of Ubuntu based on the idea that Canonical is not giving the same amount of care to Kubuntu as other flavors of Ubuntu. It’s not true because Canonical has given much more care to Kubuntu than many other flavours. But all those who believe in this myth may now rejoice as the stepchild is moving out and going to share a flat with its much loved siblings \o/

About these ads

43 thoughts on “How Kubuntu Did Not Change”

  1. I there is a kernel of truth that can be expanded on with a small change in terminology to be more expansive in its impact. I’ve taken the liberty to write one version of the more expansive truism below:

    “The detanglement allows us to move even closer to _upstream project_ without having to worry about conflicting interests, as what is good for _upstream project_ is not necessarily what is good for _downstream business interest_.”

    Will we see further detanglement between upstream project interests and Canonical interests in the future?

    -jef

    1. I am not working for Canonical, so I do not think I am fit to answer this question in particular.

      What I meant to communicate is that naturally there are differences in expectations and vision between upstream and downstream. For any upstream and any downstream (well, unless upstream == downstream I suppose ;)). Kubuntu traditionally sees itself very close to KDE and always tries to deliver the best and truest experience of KDE technology. At the same time however it is part of the Canonical support portfolio so their requirements and wishes need to be taken into account. With the latter party falling out of the equation Kubuntu has more freedom to follow upstream’s lead and vision, while only having to stay true to our own values and ideas.

  2. I’d love to see that list of 25 grow a bit bigger, but honestly- the wiki pages for the different “how can I help” sections are a bit small.

    I tried to help with the release announcements last release and it was really hard to communicate and find out who I needed to get in touch with to actually help. My beta notes got in, but I intended friendlier notes for the full release and what ended up being the actual release announcement was mostly what I’d helped write for the beta (the beta I wanted to be friendly-technical; the release friendly-average).

    I’d love to see a move towards infrastructures and policies that help attract help, rather than discourage it.

    I’m sorry about the bad press canonical’s move is giving the project, and I will continue to be a Kubuntu guy, as I type this using Kubuntu wearing my Kubuntu logo shirt.

    1. :)

      Kubuntu has a highly group driven work attitude, so it is unfortunately sometimes tricky to pinpoint someone directly responsible (that is: a person one would want to talk to if one wants to help with a specific task). In your specific case I believe claydoh is the most involved member, so he’d be the person to talk to I guess.
      What always works is spamming the kubuntu-devel mailing list though ;)

  3. If that makes copying Canonical’s “we ship beta software that sucks and don’t update it because we are not rolling release” philosophy stop in Kubuntu, this is a good move.

      1. Replacing Xine with GStreamer … sure, Xine is considered “deprecated” but Phonon-GSt was not ready that time and I still cannot hear music properly on my Natty as Amarok happily skips tracks at will and cannot seek MP3s. Were there updates to that? NO! In 11.10 then I could at least seek MP3s and in 12.04 the amount of tracks skipped goes to zero but still, that sucked.
        In 11.04 a visually broken version of Plasma NM was shipped – were there updates? No! I manually re-compiled it from latest git (which I regularly do because you don’t ship updates…)
        The same mess with Pulseaudio … just because Ubuntu uses it,doesn’t mean we need to instantly adapt in Kubuntu at a time where KDE was not ready for this. Now I don’t want to miss it anymore but still, the first thing I had to do to make my USB headset work back then was sudo apt-get remove pulseaudio …
        So I beg, please do NOT replace Kopete by KDE Telepathy in 12.04, as I KNOW you will screw it up and have users end up with Telepathy 0.3 or so with no chance of getting an update except for some PPA by a third party or waiting for 12.10.
        I could add a lot of additional cases to that list but I think you get the point. Shipping Beta software may be okay, and I am also a guy that looooooves the newest of the newest but how should I explain to my customers that suddenly $feature is gone and replaced by $brokenfeature and I can’t tell them to wait half another year to get it fixed. And I am definitly not willing to visit them each month and manually git-pull and re-compile the broken stuff. I think I will eventually have to learn how to set up a DEB server so I can have my own package source and pack recent software in that and have those packages shipped to my customers.

        1. - Phonon Xine vs. Phonon Gstreamer (not beta software)
          It is not a question of deprecated, but maintained, shipping unmaintained software is a security risk and forbidden by various policies. That is not the reason for this though, it was request by upstream (that is KDE). As it happens I am actually the immediately responsible upstream person for said request, which was directed at every and all distributions, so if you wish I can give you a complete set of reasons why exactly that switch was necessary when it was done. Also mind that the majority of your issues are a whole stack of issues in Amarok, and leaky abstraction in Phonon Xine to support the stack of issues.

          - Plasma NM
          http://cdn.omgubuntu.co.uk/wp-content/uploads/2011/02/kubuntu-natty-desktop.png looks just fine to me? Also

          - Pulseaudio (not beta software)
          Kubuntu did not adopt Pulse until 10.10, which was like 4 releases *after* Ubuntu and the release just after an LTS so the obviously best time to introduce such a change to make sure it works flawless in time for the next LTS. I’d really not say that 4 releases after Ubuntu, which also was like 1 year after Fedora adopted it, was instant really. In fact at that point >90% of the issue Ubuntu had were fixed and only some side cases needed addressing. At large the Pulse transition was very smooth actually.

          - Kopete vs. Telepathy (???)
          Upstream does not consider that a good idea yet, we do not consider it a good idea yet, anyone who remembers that 12.04 gets 5 years of support will know that it is a bad idea without even knowing how buggy both Kopete and Telepathy are. Funny you should come up with that idea.

          A word of advice from a QA POV: you really should not make your customers upgrade their systems every 6 months for the sake of having the latest software, particularly without having first assessed all regressions that might affect their business. To that extent even with a 6 month upgrade cycle I personally would still want to lag one release behind the current one, by then you do not even have to spend time looking for regressions as hopefully all are reported. Given that the average release is supported for 3 years you can do that 6 times (!) before having to choose the least terrible evil for an upgrade and to conduct the upgrade you simply upgrade your way through all releases in between. Now this iterative upgrading to the target version is of course a bit of a hassle, which is exactly why there are LTS releases. Every 2 years you get a release with specific focus on reliability with 5 years of support, so you can even skip one of those if necessary…

          BTW, having an overlay archive with additional fixes/software is a very good idea, regardless of the degree of pre-upgrade QA!

          1. In fact, in Fedora KDE, we adopted PulseAudio at the same time the rest of Fedora did, which was in Fedora 8. That was a KDE-3-based release, which means the default sound system on the KDE spin was actually aRts on top of PulseAudio. (I can assure you that it worked!) Starting with Fedora 9, we shipped KDE 4, which meant Phonon, and since then the “aRts on top of PulseAudio” hack has been used only for legacy KDE 3 applications (and these days, aRts is of course no longer installed or run by default). Early releases of Phonon did not have PulseAudio integration though, so PulseAudio showed up as one device. What we did is that I patched Phonon to give that device the highest priority (which was an effective way to make PulseAudio support work by default). Nowadays, PulseAudio support in Phonon is much better and we no longer have to patch Phonon, plus as a user you no longer have to choose between PulseAudio and being able to control device priorities. The coolest feature is that the device priorities you set in Phonon are enforced session-wide by PulseAudio, which means you actually get BETTER support for controlling device priorities with PulseAudio than without. Thanks to Colin Guthrie and the rest of the Phonon team for their awesome work!

            So I also don’t understand the complaints about Kubuntu having picked up PulseAudio “too early”, in fact we Fedora KDE packagers were complaining all that time that they were picking it up too late. ;-)

        2. Who compiles KDE from git?
          These guys offer the latest KDE in their official PPA usually a week or two after a KDE release. If you can’t wait 2 weeks… you sound like a perfect candidate for Gentoo, imho.

        3. just in reply to your last sentence.. this sounds great.. you should definitely set up a ppa for that… or better.. join the crew and be nr. 26 in the kubuntu team!!

    1. I’m not sure what you mean by that. Almost since Ubuntu started there have been non-Canonical developers with access to uploading the entire archive.

  4. It’s like in politics : Changes mean regression …
    But as we don’t pay may we expect as good as those who pay ?
    Open source usually do this at least and much more…
    Times are a changing ( but unfortunatly bad)

  5. Just for the record, Edubuntu and Xubuntu applied for LTS as well. Being LTS is not exclusively a matter of being supported by Canonical (although it certainly helps a great deal), it’s the Ubuntu Technical Board which approves this. If there is credible proof that a community wants to, and can support a stable Kubuntu release for 3 years, then it can be LTS just as Ubuntu itself.

  6. I hoped Canonical will actually drop Unity and switch to KDE, so I found this decision quite surprising. Unity is broken by design, because you will always suffer from lower performance in games, because you can’t suspend compositions without destroying your desktop. Kubuntu also offers sane user experience, so you don’t have to search for application every time in messed up Unity menu. I hope Kubuntu will become even better without Canonical support.

    1. Unity is Mark Shuttleworth’s baby, so this was never an option. However, Qt is as default in Ubuntu now, so hopefully more Unity apps will be Qt based instead of GTK based in the future.

  7. I just wanted to make clear that Canonical will not be providing commercial support for Kubuntu 12.04 LTS; the support team will be only supporting Ubuntu 12.04 LTS.

    1. I hope the support team is taking this opportunity to ask themselves the question “what do our potential and existing customers understand our support product story to be.”

      I find Canonical’s support story to be very confusing personally. But I also realize I most likely have a somewhat different point of view than the bulk of potential and existing customers. So having me go into great detail, uninvited, is not particularly helpful to anyone. But I do hope someone is taking the time to make sure the target audience knows the what’s what moving forward.

      The fact that you felt you had to clarify here without citing a documented support offering broadside or whitepaper, raises a little alarm in my head. It should be clear from citable documented support offerings what is and is not supported. It’s understandable if those documents don’t yet exist for 12.04 (and I’ve hit the sleep button on the mental alarm clock.) But man is it going to be confusing to a lot of people if an updated support story don’t exist come LTS release.

      -jef

        1. Jono, you all are confusing the community. The is a reason Mint is gaining popularity. It seems like canonical has lost the plot. For a long time it seemed like the Ubuntu team had it’s hand on the pulse of the linux community. But recently it seems like it’s floundering trying to find a way to stay afloat and like the team is panicking about profitability. Which is a fine thing to worry about, but come clean with us and do it rationally. Starting Unity was NOT the way to deal with that. Ubuntu succeeded because it stood on the shoulders of giants, as opposed to so many other distros that try to be “giants from scratch”. Starting your own DE is so much worse than even that.

          You all need to prove to us that you’re on the right track and that you have a REAL vision not just a hallucination, cause you’ve lost the faith of the community. I’m sorry, it pains me to say this I had a great time on early Ubuntu but since this Unity thing started I’ve been on edge and now I’m finally out of denial. It’s time you guys get out of denial too. :-/

  8. It means that Kubuntu has been pooped on by canonical for years and that it is not finally being kicked out of the house. Cannonical never cared about KDE despite the fact that KDE/Qt would have been much better technologies to base unity off of. Oh well, screw you too canonical. Guess there’s no need to keep looking back at Kubuntu. Hello Fedora/Chakra!

    1. I think you are being a little unfair here.

      Firstly, if Canonical didn’t care about KDE and Kubuntu why would it hire staff to work on it, produce thousands of free Kubuntu CDs to give away, and provide the infrastructure and facilities for the Kubuntu project to operate.

      All that has changed is that Canonical is focusing its resources more specifically on Ubuntu when it comes to staff time and commercial services. All start-ups need to conserve their resources where appropriate.

      Even if Canonical does no longer fund Jonathan’s excellent work, why do you feel the Kubuntu project is now dead? Kubuntu is alive and kicking…but it needs more community members to help…could that be you?

      1. The story is simple – Canonical stopped sponsoring a developer to work on Kubuntu for business reasons. The fact that you no longer offer the 3 year support on Kubuntu seems to align with that. Of course it isn’t NICE for the KDE folks but hey, the KDE community creates Free and community/openly developed versions of Canonical products (Unity, UbuntuOne etc) so it doesn’t make sense to support them… I don’t think Canonical has to explain themselves or is doing something evil in this area.

        Oh on a side note – I wouldn’t call a 500+ people company of over 5 years old a ‘start-up’ ;-)

      2. Jono,

        You really need to do better. Kubuntu 12.04 was indeed supposed to be an LTS release. In fact, Jonathan himself said so recently. There are so many references to the fact that 12.04 would be an LTS release that it is hillarious to pretend that this was just a big misunderstanding.

        You cannot even imagine how much damage this is doing to Canonical. We were set to standardize on Ubuntu Servers and Kubuntu desktops and planned to get support from Canonical.

        Now that is all on hold because we have tested and trained on Kubuntu as that is what our users and management prefers. Admittedly, we are “only” a company with about 100 employees, but I am sure there are many others in a similar situation.

        Unless Canonical very quickly rectifies this, we will be looking for support from a different Linux company. And we do, we will not be back for a long time because we do feel that roadmaps have been changed without any kind of consideration for your existing or potential customers.

        1. Kubuntu 12.04 *is* going to be an LTS release but the LTS process for flavors does not mean “commercially supported by Canonical” or that “Canonical provides security updates”. Instead the community instead provides this level of support for security updates.

          he Kubuntu team applied to the Technical Board for LTS approval, and it was granted. See https://lists.ubuntu.com/archives/ubuntu-devel-announce/2012-January/000922.html for details.

          I am sorry to hear that this is causing headaches for your company. If you would like to discuss your options further with Canonical as a support partner, feel free to email me at jono@ubuntu.com and I can put you in touch with our commercial services manager who can see what we can do.

          1. Jono,

            Thanks for the offer. I am taking this up with the rest of the C-level people at my company to see how we go forward. We will test Kubuntu 12.04 when it is officially out and decide then what level of support we require.

            If we do feel that support is needed, we will be in touch.

            Thanks again.

      3. Indeed, the poster above is being unfair. I’m not aware of any company ever offering a paid support offering to provide customer support for Fedora’s KDE. So the last sentence is entirely irrational and reads as a petty jab targetted to raise your hackles.

        The loss of Canonical’s backing as a paid support provider does not invalidate Kubuntu as a technical offering nor as a worthwhile endeavor deserving of contribution. And there is no evidence whatsoever to suggest that a team of non-Canonical paid staffing will be unable to keep Kubuntu rolling along as long as the Canonical run infrastructure they rely on continues to be there.

        At worst it means people running businesses, and were looking to standardize on Kubuntu in business deployments might decide otherwise now because of a lack of a paid support offering. But anyone who seriously factors paid-support into their desktop deployment scenario won’t be considering Fedora to begin with. More likely they’ll be looking at Suse now for a for-pay supported KDE offering.

        -jef”though I have to wonder how old and how big does a company have to get so that calling it a start-up sounds like remorse feeled nostagia instead of a statement of fact”spaleta

  9. First of all, Kubuntu is known as the “red headed” stepchild of the Buntu family and whether you believe its a myth is irrelevant.

    Most people will refer to this quote: “”I believe that the KDE community does phenomenal work, and having a community-driven distribution to showcase that work will help attract users and developers to the project. Our overall goal in the Ubuntu project is to further the adoption of free software on the desktop and the server, and we recognise that KDE is an essential part of the mix of desktop environments that allows people to find the best environment for their needs.”” from 2006 from Shuttleworth

    Read that last line again.

    Getting people over to Linux is about finding them a destkop THEY feel comfortable and when wooing developers this was always pushed forward. True or not, KDE always felt as an afterthought (and considering it was always a mediocre version up until v10.04, it didnt really matter much).
    But this business decision is about consolidating a message rather than Jonathans million dollar salary.
    Choice is something we cherish but makes marketing BS harder to spread. Keeping it simple helps keep the focus on the one product. Offering an official alternative gives the impression that “The One” is maybe not perfect. And if you let that little crack open, who knows, the user…sorry, the consumer might do something unthinkable like try another distro.
    And worse, like so many before will realize that its really not that different (how many newbies do you know that are surprised that the software is always the same no matter the distro?)

    You are right, Kubuntu will have to deal with Jon’s reduced hours to fill up (which I am sure he will still shepherd) but this move might give it more flexibility..
    Problem is perception. The same perception that called Kubuntu the red headed stepchild of the Buntus will play this newfound freedom in a different light. Many users and non-users will only know the lede and Kubuntu Is Dead will become the newest meme du jour.
    Unless Jono or some other handlers start trolling forums in favour of Kubuntu as well, trying to clam up nicely the people who are saying the same things he used to in his youth.

    Perception is a hard thing to grasp. Why do some distros like Mint now and PCLinuxOS before that raise so meteorically? THe perception that they listen to users. Thats the wonderful thing about being the underdog. Once you become ‘the man’, you are seen more often as being unwieldly while you see yourself as ‘taking the hard decisions for the good of the user’.
    (shades of we know better than you what you want.)

    Outside out own little circle jerks, these perceptions will get flamed by Kubuntu lovers who think that their ‘precious’ will somehow be diminished because it doesnt have the: blessing of Pope Mark, KDE lovers who are pissed that ‘their’ desktop is being shunted again, KDE users from other distros who believe that the failiure of an ‘enemy’ will help theirs and users from other desktops who think the same way..

    The blessing in all of this is a realisation my eldest son made a few years ago and that many users realize after theyve used Linux a a few years: using your favorite desktop is almost the same no matter the distro and going from one to the other is very easy.
    Because very often the most important decision a user makes is NOT the distro choice but rather the desktop choice. And no matter the choice, the user is ALWAYS right.
    Because the first thing you learn in business is: the user-consumer is always right.

    And as long as they use KDE, it makes really no difference which distro they use (ok, ok, id rather staple my testes to my leg than use SUSE!).

    But as a free software advocate, as long as they use Gnu-Linux, its a win for us, no matter the desktop.

  10. Hi guys,
    I’m just one of those elusive nightbirds – an enduser – AND non-techie!
    But for years and years I was a Linux advocate and mercilessly attacked in those early days for insisting that the real world wanted a GUI, not the black screen of terror.
    Occasionally, I still dip into the Linux maelstrom and think how BillG must be laughing all the way to the Bank and can focus on his big project: population reduction by any means.

    Nothing seems to have changed.
    There are three target groups out there and each needs are different:
    1. The 90%. Simplify it. We don’t need geek/sysop stuff in there at all. Look at that fruit-machine!
    2. The 8%. Commercial users of 100+ bums.
    3. The 2% Software fanatics – envelope stretchers – who don’t want it perfect, but want a challenge. (Without whom nothing would ever happen! )

    All that is needed is a Free copy of the M$Dos system – or better still these days – a BASIC Apple system.
    I believe (as an old, old marketer) that someone could do a Jobs job on Apple like he did to others and simply copy everything Apple does right down to colours and let ‘em sue (just like Jobs did).

    Won’t happen of course.
    But if you programmers could only see that all the basic users need is NOT SO MANY CHOICES!
    Aim for a Mint type idea of ….. just plug it in and all the (familiar) basics are RTG. Choices and addons come much later.

    But, hey guys, thanks for all the free fish over the years,
    Himagain

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s