#4: Plan, execute, DELIVER

Wednesday, December 27th, 2006

This is one post in a series, describing challenges we need to overcome to make free software ubiquitous on the desktop.

We are a somewhat chaotic crowd, the software libre army. Thousands of projects (hundreds of thousands, if you consider Sourceforge as a reference point). Hundreds of thousands of contributing developers from virtually every country and timezone. We are a very loosely coupled bunch.

To a certain extent this loose coupling is beneficial. Work goes on in one part of the free software universe entirely oblivious to work elsewhere, despite the fact that both pieces of work will ultimately land up on an Ubuntu disk. Keeping everything orthogonal is very nice – very UNIX. It means that people don’t have to keep too much in their heads. And that’s worked well.

But sometimes I wish it were easier to keep track of changes and have a slightly clearer view of progress across that whole galaxy. For example, it would be nice at the beginning of an Ubuntu release cycle to have a really confident picture of which projects will produce stable releases during those few months when we can incorporate new upstream versions. It would be even better if, during the release cycle, we knew immediately if there was a *change* in what was going to be released.

This predictability is one of the major reasons we picked Gnome for the first Ubuntu desktop release – we knew with some confidence when it would be possible to ship a clean, fresh “chunk”. And Gnome has been a superb partner in that.

This coordination could go deeper than simply the planning of releases, and release management. One of the big advantages I’ll bet Microsoft has is that they have a single bug tracking system for developers from Office to the Kernel, with everything in between as well. That means that all their developers can have high-bandwidth conversations about any bug in the system. Compare that with the balkanised free software world, where I have to create a Bugzilla account any time I want to work directly with a new upstream. I hope Launchpad’s bug collaboration features will make it easier to coordinate between those bugzilla’s, at least for projects that are using Launchpad or where the bug has already been filed in Bugzilla (LP can link to other bug trackers).

Bugs, feature planning, release management, translation, testing and QA… these are all areas where we need to improve the level of collaboration BETWEEN projects. I think Launchpad is a good start but there’s a long way to go before we’re in the same position that the competition is in – seamless conversations between all developers.

36 Responses to “#4: Plan, execute, DELIVER”

  1. Lunatic Says:

    well..is your launchpad going to be open sourced? This kind of bug tracking system will help the community as a whole if it’s open to public and deployed at a centralized location.

  2. S.Yasa Says:

    Considering a central bug tracker on all software related to ubuntu would be nice. However, one of the main ingredients to materialize such a goal is to have a peer to peer like database of bugs. No matter which bug tracker tool is used, searching for a particular bug will query all network of peers. Imagine Launchpad providing bugs from Bugzilla and vice versa.
    Never theless such a feature would need a specially crafted format to keep track of bug converstations and status flag.

  3. feranick Says:


    All nice and desirable. However I see that Ubuntu would benefit as more attention would go into QA. There are lots lots of little details (not to call them bugs) that go through the release process that diminish the pleasure to use ubuntu itself. If, say, 7.04 would have the same features of 6.10 but with lots of bug fixing, I would be more than happy. Going after new features, both in terms of planning and release at cost of reliability and stability is to me a big mistake, which often Microsoft seems to be doing in their products. I hope Ubuntu won’t follow the same path.


  4. Jonathan Carter Says:

    Mark, I’m not going to sing the ‘Open up Launchpad’ song again, it’s already a controversial and huge issue in the Ubuntu community. However, it may be beneficial to the open source community to open up some parts of Launchpad that will not hurt your business plan. For example, the most horrible part of Launchpad, which is pointed out regularly, is the user interface. Quite often, I stumble upon things that could quite easily be changed that would make quite significant UI improvements, sometimes as simple as a small change in the Launchpad style sheets, yet, all I can do is file a bug, and the Launchpad developers seem to be working so hard on their big hairy audacious goals, that they don’t have much time for the smaller, seemingly insignificant bugs that clutter the Launchpad.

    If Launchpad’s UI could be released under a free software license, it will already be a benefit to the ‘open source community’, at the very least, the Ubuntu community. Launchpad has already improved ‘realtime collaboration’, but Launchpad itself is in need of some help, and I think it’s time that you involve the community some more in achieving a better Launchpad.

  5. Jon Says:

    As per Launchpad going open source, I don’t know that there is enough enertia behind it to garner a rally of development troops to work on it. I don’t know if the intent is based in the “secret sauce” camp, but I’m not an insider nor even a heavy hitter. I use ubuntu as a desktop and for some enterprise work in my office but remain aloof of the politics of open source.

    Jonathan has some good point, the UI does need work. The back-end seems stable and everyone of my guys who browse it love it. Bugzilla has a huge penetration but only because it’s the only oasis in the desert of quality tools.

    I’d like to see Launchpad go open source if it makes sense for the open source community. From the perspective that the community lacks a tool as powerful as Launchpad, it makes sense. But given the size and complexity of Launchpad I’m not so sure unleashing it wholesale to the community is wise either.

  6. Larry S. Stotler Says:

    I’ve pointed this out I don’t know how many times before, but no one seems to care about making it happen. Here’s the(IMHO) biggest thing keeping adoption of Linux from happening for the average user:


    WinDoZe wins out for 2 reasons:

    1. The biggest reason is that it comes preloaded on just about every new PC sold. That’s Micro$oft’s is still able to force PC manufacturers to have WinDoZe preloaded, that’s what people are going to use.

    2. When someone buys a piece of software or downloads a program like Firefox from the net, when they have a WIndows system, it installs on just about every version of Windows from 98SE forward. While it’s true that more and more apps are requiring 2000/XP, it’s still EASY for them.

    With Linux, you have a different version of each program for each distrobution. Then you have a different file for each version of each distro. OR, you can install from source and run into problems with dependencies. While Debian/Ubuntu have apt-get, that’s not good enough. We need a system that will allow programs to install on any distro without difficulty. The Linux Standard Base was supposed to be a start, but it has sputtered out. Issues like one distro using /usr/lib and another using /usr/local/lib need to be sorted out between the major distro venders. At that point all the clones will get it too.

    If you want to truly help, let’s work for a Universal Package System. Instead of trying to steal developers from other projects.

  7. james Says:

    so, you’d basically want to change the open source model to be more like microsoft’s? Keeping launchpad closed source is a good start in that I guess.

    Actually this is more like removing the volunteerist attributes of open source by requiring a fixed release schedule (we do have other work and lives, some of us are in it for the code, not the money, sorry mark)

  8. Eugenia’s rants and thoughts :: OS != Distro :: December :: 2006 Says:

    […] There is nothing better than seeing people “getting it” as they go along. Mark Shuttleworth of Ubuntu is one of these people. In his latest blog post he discusses that projects must co-operate and interoperate with each other, and one first step to do that is to have common release schedules and bugzillas. He is so right about this. The No1 thing that makes GNU/Linux “clunky” and “difficult” is the fact that each project does its own thing, without architecting their software in a way that plays well with a grander scheme — because there is no grand scheme as a whole. This is the reason why Gnome/KDE feel so “cut off” from the underpinnings of the OS they run on. But Mark gets it, and he gets it because he is in the business of the OS desktop. I am sure that Red Hat and Novell people also got that in the past, but their primary business were always the server market, and therefore never cared to comment or pursue a grand plan of transforming a “distro” (as in collection of random software) to an “operating system” (as in one logical entity that was engineered to interoperate perfectly). However, I am sure that when Mark and others in Ubuntu came together to create Ubuntu Linux they didn’t have such ideas but they realized the needs of a modern desktop engineering as they went along. […]

  9. RMX Says:

    If this article is true, it sounds like what we really need is an open-source competitor to Launchpad; in much the same way the kernel commuinty needed an open-source competitor to bitkeeper.

    However, I mostly think the whole premise is false; and that to “have a really confident picture of which projects will produce stable releases during those few months when we can incorporate new upstream versions” is not a desirable goal for the community at large anyway.

    Having a confident picture of the future almost by definition means absurdly conservative planning – and absurdly conservitive planning by definition means not progressing as fast as one could.

    I *WANT* the upstream projects to Release When Ready and Delay Projects When Needed and Cancel Features When Needed and Attempt-Risky-in-the-short-term-but-Beneficial-in-the-long-term changes when appropriate.
    None of those lend themselves to corporate planning or release schedules; but those are the attributes of F/OSS projects that outperform other F/OSS projects (and I’d say postgresql vs mysql is a reasonable example fo this).

    I think the best thing a commercial release should do is to not try to predict the future of upstream projects — and especially not try to dictate the future of upstream projects — but rather focus on the already released versions because those are the parts where you really can make high-confidence decisions about the stability and release-worthyness of a project.

    Allowing the upstream projects to move faster (at the risk of having a fragile build because they’re in the midst of a major update) is a good thing; and release planning shouldn’t get in the way of this.

  10. Dusty Says:

    open source projects should all get synchronised and use launchpad and all live in the happy ubuntu tree?
    You’ll excuse me if i dont follow the ubuntu sheep….

    This is just some more proof mark is a moron who doesn’t get it.

  11. Evanc Says:

    Up-to-date with Technology – Not Necessarily Up-to-date with Life!

    Hi Mark,

    I often read your postings, informative, interesting and quite technologically insightful, at least for me. The later maybe even a little too much, at the expense of thinking/talking about things which are equally or more important.

    And I suppose it is normal for everyone to see life through the prism of what is meant to be their ‘profession’, in your case technology, and then look outwards, i.e. towards education, innovation, virtual/real world or ‘humanity to others’.

    That is perhaps what also gives you a sense of making the best use of what you have, contributing towards what you feel you can contribute best i.e. technology.

    Having said this, it is only logical what you have done, making the blog of Mark Shuttleworth be a blog of Ubuntu, Open Source or whatever is the latest technologically edgy project that excites you and you want to promote.

    As for my opinion on the last, it seems to me that Ubuntu and the conferences organized around it, in terms of substantial innovation, are a little bit like the ‘the king’s new clothes’. Sometimes it appears you have a tendency to combine existing ideas, and convince yourself that you have come up with something greatly new. So in this sense I prefer your promotion of innovation to your deeds on it.

    But that is not extremely important and not my point. Since it is all work in progress, for me it is the idea and purpose behind it that counts most, and the same as for most of your other projects, these are based on ‘humanitarian’ principles. I think not many people are ready to put the life of humanity before their own lives. Yet, it has proven that people who have satisfied some basic life needs are more capable of such humanitarian aims. That might be an example, and a faithful one, that behind each self-interested human lies a humanitarian to be triggered. (Although, hopefully for some by something other than 300m.) Therefore for your underlying aims, I give you an ‘A’.

    And it is these and similar aims that perhaps I would like to see you ‘blog about’ more. Being up-to-date with technology is not necessarily being up-to-date with life!

    I can not outline distinctly the difference between being up-to-date with life and up-to-date with technology, since technology seems essential part of life, however it is a sub domain of it and it is the remaining elements of being up-to-date with life that I thought you should consider more.

    These are not about the system you are using to read my comment, not the computer, not the means you used to get to your current location. For example, these might be the obvious or subconscious reasons you would do any of the above. Reasons which when considered might yield a different outcome, even technologies, rather than when merely adventurously aiming for a higher level of the technologies above.

    In essence, not about having the tools, or the adventurous tools and gadgets that are valuable to you merely based on how unseen and fascinating they are in themselves, but rather having good reasons to develop, tools too, but not only, and maybe aligning with those foundations on more regular basis.

    I can only add that for me some of these updates with life involve checkups on the ideological and value direction that humanity is taking, or going towards, or one that it isn’t, or one that it should be.

    And looking at the people who write here, I might be might be a minority in terms of interests but hopefully not so much in terms of goals.

  12. Fat Mobil Content » xmen acrlight Shuttleworth: Open Source projects need better collaboration Says:

    […] xmen acrlight "Bugs, feature planning, release management, translation, testing and QA … these are all areas where we need to improve the level of collaboration BETWEEN projects. I think Launchpad is a good start but there’s a long way to go before we’re in the same position that the competition is in – seamless conversations between all developers." xmen cartoons the whole gangread more | digg story […]

  13. stained class Says:

    TV commercials in the United States, we need to market the name of Ubuntu if/before M$/SUSE is marketed. Unless you want people to forever think that Linux is Microsoft’s. Yes, people who consume ads in the majority are stupid, but there needs to be marketing in the U.S. on TV, NOW for Ubuntu, not “some day”.

  14. aguafuertes Says:

    I second Jonathan’s comment: the interface is very hard to use, at least for non-developers. Having Launchpad components under an open license could increase usability rapidly, letting the right people (designers, web developers) scratch their itches…

  15. Diego Says:

    Opening launchpad is not the solution. The solution is adding such feature to bugzilla!

    It’d be also nice to move towards package mainteinance in upstream (at least, in part) instead of doing it N times for every f*cking distro. The number #1 issue for inter-distro compatibility is the package namespace – x.org has a different package name in ubuntu and in fedora and makes dependency checking impossible. This would be automatically fixed if part of the packaging mainteinance would be moved to upstream (the upstream developers would set the name and the dependencies, and the number of sub-packages created for such package, etc)

  16. John Nilsson Says:


    The solution isn’t to improve one centralized authoratice service, the solution is to work on standards for interoperating between decentralized services. Design the standards that are needed for commuinicating the needed information between projects.

    I think we need to start designing a better OSS network. Do we need an authoritive namsepace for that? a P2P service? XMPP extensions for data interchange? Better network-VCS integration?

    As Diego suggest can we push the package maintenance further upstream? Do we need new packaging formats for that? Is the trusted tar.gz the best format? Do connary provide som exiting possibillities?

  17. John Drinkwater Says:


    Would you like to encourage the launchpad developers to add OpenID support into the site?

    It would give all those people coming to launchpad without an account an incentive to report bugs, because, after all, needing to register for yet another (launchpad.net in this case) bug tracker is a pain.

  18. Ant Bryan Says:

    You should read this story about the coordination within Microsoft: http://moishelettvin.blogspot.com/2006/11/windows-shutdown-crapfest.html

    “The periodicity of integration decayed exponentially and unpredictably as you approached the root so it ended up that it took between 1 and 3 months for my code to get to the root node, and some multiple of that for it to reach the other nodes…

    So in addition to the above problems with decision-making, each team had no idea what the other team was actually doing until it had been done for weeks.”

  19. Oliver Herold Says:

    Open source is chaotic because you have the free choice of doing what you want.

  20. Jeff Says:

    Little off the subject but when it comes to unifying I think the largest but most beneficial undertaking someone like Mark could champion would be to build a new desktop which combines the best of KDE and Gnome. This kind of unification would be an enormous driving force in promoting linux on the dektop. This has to start at the core and would take a ton of resources but if it could be pulled off it would be legendary!

  21. Adam Dutko Says:

    Why doesn’t someone create an “RSS” feed style meta-bug tracking system that provides a universal view of all the bug trackers for each open source project in one place? It would simply require the registration of an account and the url for each monitored bug tracker: it might even be simpler than that… I’d take a stab at it, but I’m busy with other projects. Just a thought.

  22. Drew Says:

    I think Launchpad is missing the point personally. Bug tracking software only helps with coding problems and does nothing for usability issues. At the moment there is nowhere to give feedback on such issues, and you get publically attacked for doing so as people see it as an attack on Linux and thus themselves.

    Ubuntu’s GUI has barely changed since it’s first release 2 years ago, and much as I hate to say it, its practically a clone of the Windows GUI only split across two bars. Since there is no real public comment on the UI, and no way of reporting and submitting suggestions (that is worthwhile) the back end quality of the code is fantastic but the usability is often poor and erratic.

    Most of my ideas and criticisms of Ubuntu (and Linux) are about the GUI and HCI yet I feel I have nowhere to share my thoughts (as do, I am sure, many other people) as they do not fall into the category marked ‘bugs’ and you get treated like a troll if you try to discuss such things on a forum. It’s this that needs to be addressed rather than pinning all hopes on yet another bug-tracking solution.

  23. Erik Says:

    That is the #1 reason perhaps why open source desktop will not really take off. The lack of coordination and a force to create and support determined push forward is critical for some of the larger goals. Any lone smaller project can produce something smaller… Now try to build us the centralized management platform for Ubuntu, something like what eDirectory or Active Directory are like. You can’t. You will not be able to.

    That sort of stuff requires a lot of competent people to push for the same goal and it is not going to happen all by itself any day near. Without centralized free management of the desktops the corporations will keep going for Windows. Lack of such features is a huge letdown for anyone that has got larger environment. The TCO of the desktops just is too high and there are too much product risks without a free extensive platform for the task.

    The tools I’m talking about should act consistent, cover the correct critical areas and just work. No “read a howto and battle for weeks” sort of stuff you see daily. It’d require integration work, re-implementating some ancient stuff, and engineering new components. A shared vision is a must there. And… Perhaps Launchpad too.

    I just wish the core Ubuntu people would stop thinking so small, step out of the box and notice there are some 5-10 areas that could make killer sellers and deliver those things. You talked more about reliability but I’d like people to deliver the impossible…

  24. Chris Ward Says:

    I live in the UK, and there’s a couple of blockages that I know of that you might be able to do something about.

    1) My professional institution, the Instituion of Engineering and Technology
    carries advertising from a Microsoft business partner co-branded with the IET logo, offering discounts on Windows and Office.
    Now, carrying advertising from any legal commercial organisation is OK; but co-branding is not OK. In my opinion it is inappropriate for a non-profit professional organisation to be recruited as part of the distribution channel for any for-profit corporation; also, it compromises my professional independence. The Microsoft business partner should not be able to ask the IET for confirmation of membership; only potential clients considering engaging my professional services should be able to do this.
    I believe other professional institutions also engage in this co-branding.
    is an on-line version of the advert in question.

    2) My children all attend schools which are Microsoft strongholds. The schools have policies of ‘no downloading software from the Internet’, and ‘no bringing software in from home'; which means they are unlikely to come across Ubuntu Linux. Further, Microsoft have a ‘Student and Teacher’ scheme, whereby with a signature from the schoolteacher I can have access to discounted prices on Windows and Office. This engages the school as part of the Microsoft distribution channel (again, a discount for a target section of the market is OK, but asking the school for a signature and being able to verify that the signature is indeed from a schoolteacher is not OK).

    I suppose a reasonable response would be to offer to supply the IET with an adequate quantiity of IET-logoed Ubuntu CDs for them to distribute with their magazine http://www.iee.org/oncomms/magazine.cfm , and also to offer to supply Ubuntu CDs for distribution with ‘teacher’ literature to schools; but I’m not sure that two wrongs would make a right.

  25. Blue Says:

    Try to begin to integrate dehs[1] data to launchpad

  26. Jason A. Bong Says:

    Cheech N Chong should make a movie about Ubuntu, what do you think?

  27. links for 2007-01-03 at rejon.org Says:

    […] Mark Shuttleworth » Blog Archive » #4: Plan, execute, DELIVER Good post on launchpad and what its about… (tags: launchpad linux ubunti shuttleworth) […]

  28. Tal Says:

    When I heard that there will be a 3D Desktop feature in the next version of Ubuntu, and that I will be able to move (literally) from one window application to another in the desktop 3D space (the mouse will function as a small spaceship in space traveling to the application icons for running FireFox for example).

    I was wandering if there will be a feature of sharing the desktop with my friends in a 3D way. I mean that I will be traveling with the mouse from my desktop to my friends desktops and exploring and playing with theirs applications?
    It will be like traveling in space but a virtual one.
    Flying with my virtual spaceship mouse from one desktop to another, finding new softwares running in other computers, meeting my friends with my spaceship mouse, finding theirs spaceship mouse spaceship coming visit my desktop to play and so on.
    The Desktop world will be the perfect virtual space of computers, connecting me with other people in the most abstract and intuitive way of traveling in virtual desktop space.

    The next step will be creating virtual cities of desktops and there I will be able to put my virtual desktop in the street of one of the available virtual cities. after that I will be able to travel in that city in the most intuitive way, meeting new neighbors desktops, visiting my friends desktops in other cities, exploring this virtual universe.
    When I will get tired I will sit in a bench in one of the virtual parks of those cities, meeting new people, talking and sharing pictures, asking how is the weather and how great it is living like that in 2012.

    Don’t you think it will be great that virtual will meet reality?

  29. Life according to Corey » Blog Archive » Open source release coordination Says:

    […] This issue comes to mind again as I read a post by Mark Shuttleworth about release coordination across the global OSS community. He is looking at the problem from a different perspective, one where they (Ubuntu) build a distribution that changes with the various new releases of included programs, and in many cases needs to be planned around the release schedules of those projects. It would help his team to be able to plan their project release around the release schedules of the various included projects. It would help my team for all of the major projects to see each other’s schedules and work to avoid release overlap. (sure, in big business you want to beat your competitors to the shelf, or at the very least release at the same time.. but big business is not relying on a network of voluntary servers and bandwidth constraints either) […]

  30. Apt Says:

    People are building Ubuntu packages that Ubuntu is not putting in apt.sources.

    Instead Ubuntu maintains horribly outdated packages of its own. Example tor, “Do not use the packages in ubuntu’s universe. They are not maintained and most likely old and therefore miss out on stability and possibly security fixes.”

    And TrueCrypt is just a no-brainer. They offer Ubuntu builds.

    All you need is an apt.sources line and the public gpg keys on Ubuntu CD. (You ask me, tor should be included directly, small and useful for “firewalled” nations. As a public service (human rights related) it would be nice of Ubuntu to run some tor exit nodes, sponsor EFF work, etc.)

    And of course, Ubuntu should come equipped with apt.sources and gpg lines to download WINE, that is an obvious one. Right now people have to jump hoops as with the others.

  31. Henri Bergius Says:

    There is good discussion about the merits of open sourcing Launchpad in:

    In my opinion, Launchpad being free is a requirement before major projects like GNOME will consider switching to it.

  32. sb Says:

    Mark, let’s pretend that Trolltech wrote an awesome bug tracker, far better than the mess that LaunchPad is turning into. Let’s say further that they were actively trying to get other projects to use it through blog posts, etc. Maybe they have hinted about the possibility of one day releasing the source but, for now, all your info will have to be stored on bugz.trolltech.com.

    Would you move all Canonical’s bug data into such a system?

    I can guess: NOT IN A MILLION YEARS. It’s not even worth talking about. We were shown by BitKeeper what can happen with this sort of philanthropy. Why would anybody risk this when they can just run their own copy of bugzilla?

    In summary, I find this blog posting confusing. I’m sure that you’re aware that no serious external project will ever move to a Canonical-hosted LaunchPad. The data is just too precious. So, are you saying that you expect other projects to move to LaunchPad despite this? Or are you now hinting that maybe LaunchPad will be open-sourced after all? A little clarification would be very helpful. Thank you.

  33. CSL: Cheli Software Libre » Interesante reflexión de Mark Shuttleworth sobre la integración de proyectos Says:

    […] Mark hace una reflexión que me ha parecido muy interesante sobre la integración que sería deseable entre proyectos de software libre. La ventaja que existe en el mundo del software libre es que desarrolladores de todas partes del planeta, de una forma aparentemente caótica desarrollan piezas de software fantásticas. Un desarrollo puede depender de otro y no pasa nada porque de ese desarrollo se espera según el plan de ruta del proyecto que tenga implementadas ciertas cosas y al final todo se integra más o menos sin problemas. […]

  34. Jucato’s Data Core » Archives » Like a river flows: Upstream and Downstream Says:

    […] would have to subscribe to each and every bug tracker. Same with downstream. Perhaps this is what Mark Shuttleworth has been trying to address through Launchpad (but that in itself has some subtle issues). Perhaps we don’t need a […]

  35. Juan Carlos Torres: Like a river flows: Upstream and Downstream // The Linux Index Says:

    […] would have to subscribe to each and every bug tracker. Same with downstream. Perhaps this is what Mark Shuttleworth has been trying to address through Launchpad (but that in itself has some subtle issues). Perhaps we don’t need a […]

  36. Mighty Linuxz » Shuttleworth: Open Source projects need better collaboration Says:

    […] read more | digg story […]