Year of the Linux Desktop

Why has there never been a Year of the Linux Desktop? Perhaps for several reasons, in no particular order.

  • No common package management. Non technical users want a simple click-and-install solution regardless of the distro they use and for any software they want to use.
  • Packages often break with each new OS release, requiring recompiling.
  • Not preinstalled by common hardware vendors.
  • Short life cycles and break neck rapid release development.
  • Unlike the server market, the home and non enterprise market is more challenging to monetize.
  • Inability to run popular Windows games.
  • Inability to run a handful of popular vertical software apps, such as Quicken, QuickBooks, Photoshop, Evernote, or AutoCAD — without the convolution of WINE.
  • The absence of a one-click solution to install non-free apps and support, such as media codecs, proprietary drivers, and firmware.
  • Peripheral support remains hit-and-miss with certain vendors.
  • Idiotic statutes such as DMCA.
  • Non technical users who are aware of Linux still perceive Linux as a geek’s operating system.
  • A persistence by Linux developers and advocates with doing things the “Unix way” — whatever that ambiguous definition might be, rather than just making things usable.
  • User hostile solutions like manually editing the boot loader with nomodeset to fix video problems.
  • Overwhelmingly designed by geeks for geeks.
  • Software patents.
  • Developer presumptions that non technical users will use a terminal to solve problems rather than produce robust GUI solutions.
  • A false belief that proprietary software is inferior in all ways.
  • Developer attitudes of “We design what we want and like,” rather than make usable and appealing software.
  • Developer attitudes that they know best rather than actually conduct usability testing with non technical users.
  • An unwillingness by developers and advocates to accommodate non technical users.
  • The NIH (Not Invented Here) attitude of developers (fragmentation) within similar software projects.
  • An RTFM attitude by developers and advocates rather than create robust GUIs.
  • Chasing OS “convergence” to the point that desktop usage is crippled beyond sense.
  • A user shift to mobile devices.

Manually installing Windows might take a couple of hours compared to 15 minutes with a Linux distro, but the opposite is true with the out-of-the-box experience after booting into the desktop. Tweak, tweak, tweak to create a desktop usable by non technical users. This is a problem more with distro maintainers trying to impress fellow geeks than the software developers.

The arguments that there are too many distros and desktop environments are without foundation. Most non technical users are not even aware Linux exists, let alone the many choices. Besides, nobody offers this argument about cars or bread toasters. This is a flawed argument speculated on by Linux advocates only.

Where to start? Choice is not an obstacle — choose a distro and a desktop environment. Configure that system for non technical users rather than for geeks. Provide support for non technical users. Stop treating systems as geek toys. Develop software that is intuitive and usable by non technical users. Stop depending on geek work-arounds, such as clunky shell scripts, terminals, and obtuse boot parameters. Empahsize GUI controls rather than scripts, terminals, and editing config files. Engage non technical users in software testing rather than singing to the choir of fellow geeks. Become involved in support forums rather than isolating support discussions to elitist ivory towered IRC chanels.

Backwards compatibility is a joke in Linux software development. Developers routinely rip support for older hardware or older tool kits. Non technical users do not understand why they can run Windows apps from 15 years ago and can’t run Linux software from only a year ago. Upstream support software often is changed to tickle the desires of developers and those changes break all existing software dependent on that software. The general developer attitude with this kind of development seems to be shoulder shrugs.

The argument that the desktop no longer matters is empty. The desktop is not going to disappear any time soon in the enterprise. Or with home users.

A note — Android and Chrome OS are not Linux based operating systems. Please stop fooling yourselves about that. Those OSs are loss leaders, developed primarily for data mining to produce targeted advertising. That goal conflicts with free/libre philosophy. That does not mean the interface is poorly designed.

By the way — anybody already using a Linux based system seldom worries about the Year of the Linux Desktop. That day and year arrived long ago.

Posted: Category: Commentary, Usability Tagged: General

Next: Youtube Autoplay

Previous: Adding a Dictionary