Ubuntu vs. End-Users

I’ve recently switched to working with GNU/Linux for most of my time, more specifically the excellent Ubuntu distribution. A few glitches aside, I’m really quite happy with the daily user experience.

And user experience is the key word here (or words, as it were): while I went through a Linux From Scratch phase, followed by a Gentoo phase, these days I don’t treat the OS I use as life’s most exciting text adventure any longer. I treat it as a tool, that has to satisfy my requirements as quickly and painlessly as possible.

Ubuntu claims to be — or to attempt to be — that tool. Mark Shuttleworth, the guy who initiated the project, clearly stated he wants to establish Linux on the desktop — that is, with “Joe Average”-type end-users.

Unfortunately, Ubuntu has a ways to go before reaching that goal, and last week I had a glimpse of where some of the problems may lie.

I ran into a very simple end-user problem scenario: I couldn’t get the OS to do what I want. In as little detail as possible, the automatic update features of the OS are generally great, and you can configure a lot about how updates are to be made, but the most sensible configuration (as far as I am concerned) is not supported. So I wanted to contact “the Ubuntu development team”1 and file this as a feature request.

Didn’t work. Launchpad, the point of interaction between end-users and developers for everything Ubuntu does not have a “feature request” section similar to how Get Satisfaction or Uservoice provide. That is, there is no place for me (in my capacity as an end-user) to file my dissatisfaction, and get a response and status updates from “Ubuntu”.

There’s a bug report section, but when I filed my feature request as a bug report, it was closed and I was referred to a mailing list. If you ask just about any non-FLOSS-developer what a mailing list is, they’ll scratch their head and think of junk mail. Clearly that wasn’t the best message to send to end-users.

The bug report was also converted to a “question”, a section where end-users can, well, ask questions on how to do stuff and presumably get answers from the community. I didn’t have a question, I had a feature request. I expect this particular “question” to hang in limbo for the rest of eternity…

  1. Which doesn’t exist, as such. There are plenty of development teams, and probably even some people taking care of the overall user experience, but not a single team responsible for everything. []

  • Andy

    I know what you mean actually about the O/S and desktop no longer being the big text adventure of your life, but now is a tool that you want to use to get shit done. I went through the same phase a while back, and was one of the reasons I gave up using Slackware on my desktop; after I had done my basic apprenticeship and learned what I wanted (and needed) to learn to bring myself to a level of proficiency, I then wanted my linux desktop to support productivity.

    I used to believe that linux on the desktop would inevitably create something the equal of windows, but now I don’t believe that. The reason Apple and Windows have the desktops they do is because they have commercialised the production work and serve much more singular goals than open source communities are capable of. Ubuntu have come the closest to creating a commercially viable desktop o/s, but even they are hamstrung because of resources they don’t have because linux on the desktop is still less than 2% of overall market share.

    I think this is a chicken and egg problem; if someone builds an end user Mac equivalent linux desktop, the linux desktop will become more popular, but until linux on the desktop becomes more popular there isn’t that much incentive to build linux desktops by commercial entities.

    And maybe this will open the doors to some flaming, but the only way a Mac equivalent desktop will be produced by the linux world will be by a commercial entity. A pure community driven open source group won’t do it; group consensus and fractured goals aren’t going to cut it in the long run. Gnome is experiencing this right now. Yes, Gnome and KDE are very good, but they still don’t stack up the same as Apple’s desktop releases.

    Andy.

    • http://www.unwesen.de/ unwesen

      Well, GNOME seeks to emulate Apple, and KDE seeks to emulate Windows. Both have surpassed what they want to emulate in some rather technical aspects, but in terms of usability, they both lag behind.

      Actually, many people would argue that. The problem, as far as I am concerned, is that perspective matters: a KDE or GNOME developer will find it very easy to point to awesome stuff their team has built, because there’s plenty of it. However, little of that awesome stuff matters to end-users.

      To give an example, WinMac sports keyboard shortcuts for closing tabs, windows and applications (to name the ones I use most), that are supported by just about every application out there.

      Contrast that to my current Ubuntu, and some apps are closed via CTRL+Q, others via SHIFT+CTRL+Q, etc.

      The reason is that, as you say, there’s no commercial entity that pushes development in the right direction (for a definition of “right”). In this case, there’s no entity that can enforce the desired behaviour in all applications that ship with the platform, or are sold by the same company at least. Third-party downloads don’t count, they don’t always behave well on WinMac either.

      I do question that the entity behind this sort of push needs to be commercial, though. I strongly believe it’s possible to have a non-commercial entity here. It’s a hard sell, because no such entity currently exists – but if you compare to FSF or ASF, both enforce stuff across a wide variety of software projects with great success; the stuff they enforce just isn’t UX related.

      On the other hand, a commercial entity might have the easiest time unifying developers’ goals. As long as someone pays your wages, you don’t tend to argue with them quite as much as if they try to influence what you do in your spare time. Setting up a non-profit that can reliably pay developer’s wages for the years it takes to develop a good desktop experience is going to be trickier, I’m sure.