Possible futures for the Linux desktop – the full list from LinuxCon

Print Friendly

On Monday at LinuxCon 2009 I gave ten possible futures for the Linux desktop. Several articles have highlighted one or two or these, but here is the full list.

Of course I added a few spoken comments, but be certain of one thing: the Linux desktop is not to be discounted and, as I list as the fifth possibility, it could end up with sizable marketshare:

  1. It goes away.
  2. We stop using desktops, so who cares?
  3. The Linux desktop becomes a tactic instead of a strategy.
  4. One Linux desktop distribution ends up with 90% marketshare among those using Linux desktops.
  5. One Linux desktop distribution ends up with 90% marketshare among all desktops.
  6. We reach 33% / 33% / 33% parity with Microsoft® Windows® / Apple® Mac OS® / Linux, plus or minus.
  7. We stop pretending that it will be a drop-in replacement for the dominant desktop operating system, and make it something better.
  8. The enterprise sweet spot for Linux desktops is virtualized Linux desktops.
  9. We focus on usability, stability, security, reliability, performance, with some cool thrown in.
  10. It’s the browser, stupid.

Also see: Life with Linux: The series


  1. From the perspective of the customer the desktop is the least of its problems they want applications that are capable to perform the tasks the way they expect them.
    This have been the demise of the linux desktop and that is why Apple is gaining. people used linux and are less favourable with Microsoft but because they need certain applications they go with Apple instead (yeah its also pretty nice and good looking).
    When web applications like Raven and Adobe Air applications will dominate the office email/social networking and others we will see the proliferation of Linux due to its performance and stability.
    So i will take 10 if i may.

  2. 11. It will be as pretty, cool, easy and nice to use as a Mac

    Asking too much ?

  3. Our company has been deploying Linux clients as our standard desktops since 2004-2005, so with the majority of our desktops now running Linux we may be able to provide some relevant detail:

    1. For us it was the applications, not the desktop flavour that was of the highest importance to the end users. We started with browser delivery where possible and appropriate (eg Domino Web Access, Pentaho, dotProject, TYPO3), and then standardised on as many cross-platform “open” applications as possible for the remainder (eg OpenOffice, Firefox, GIMP, Inkscape). Luckily our core ERP application was accessed via SSH, so again effectively a “clientless” deployment.

    2. The browser is king. Newer applications such as Google Apps only reinforce the original strategy of deploying browser based Apps where possible.

    3. We have used wine at times to deploy some “must have” windows apps locally, and we also utilise a small number of server based Windows VMs for “essential but finicky” windows apps.

    In summary, the Linux desktop IS viable if you lay out a migration strategy, but don’t expect end users to be comfortable with a “big bang” upgrade (and associated learning curve) if you try and migrate Apps and the desktop OS all at once.

  4. For us, cost and ease of management made GNU/Linux the choice. It was so easy to implement a GNU/Linux terminal server cluster with openLDAP that we had money left over to buy other hardware. We did not need to migrate any apps. We use mostly what comes standard with a GNU/Linux distro plus web interfaces to necessary databases. Our cost per seat was less than half what going with Vista would have been and our cost of maintenance almost disappeared. Users loved the performance. Administrators loved the ease of management.

Comments are closed