Tuesday, December 23, 2008

Ubuntu Usability

I've been experimenting with Ubuntu for a few months now, and it is a great operating system. With netbooks on the rise, and increased notoriety, more and more average consumers are trying it out. As far as it has come, there are still problems with hardware and software incompatibility that will keep many users away. Putting technical issues aside, I wondered how usable Ubuntu was for an average Windows user. Usability could play a major role in the adoption rate of the Ubuntu OS. I worked with a team of my fellow Bentley students to perform a usability test on the operating system.

Six individuals participated in a usability test of Ubuntu version 8.04.1. The test aimed to assess Ubuntu’s usability and identify barriers that might hinder Windows users from converting.

Participants completed some tasks with relative ease (reorganizing files, using the word processing application). However, participants struggled to complete other tasks (connecting to a wireless network, changing the screen resolution).

A few participants reported interest in using Ubuntu, others hesitated to abandon their current Windows operating system and voiced concerns about application incompatibility and leaving their comfort zone.

This report presents the project’s goals, our usability test methodology, and our findings and associated recommendations.
Ubuntu Usability Test Report
View SlideShare presentation or Upload your own. (tags: usability ux)

UPDATE: Original PowerPoint Presentation: View ubuntu-testreport-final


Anonymous said...

I am glad that someone is taking usability seriously. Too bad that this has not been done until now. I think that more participants are necessary, but I hope they will take into account the recommendations.

Anonymous said...

I'm also glad that you are serious about usability. Unfortunately about 30% of this work is now meaningless for just one reason: the test is about Ubuntu Hardi, but the new version, Ubuntu Intrepid has a new Network manager which looks completely different (and I'm sure is usable like never before).

Why did you take Hardi on November, when you could take Intrepid?

Vadim Peretokin said...

@anonymous1: I'd take your words back, because it has been done before: http://www.betterdesktop.org/wiki/index.php?title=Main

... and is an ongoing effort still: http://arstechnica.com/news.ars/post/20080911-canonical-to-fund-upstream-linux-usability-improvements.html

I'll see if I can get those guys to look at this study though, it is well done and I agree with some points.

Fitech said...

Great to hear feedback.
@anonymous1 More participants would be nice but 5-7 is often enough to uncover many of the high priority usability issues (see Virzi)
@anonymous2 Yes, we planned the study before Intrepid was released, and I wish we had tested the latest version. I have even seen better networking functionality on different computers running Hardy Heron depending on the networking hardware support.
@vadim Thanks for the links and the support. Very interesting.

technoshaun said...

I would like to see more studies done myself.

Such studies can help us improve the interfaces and make the GUIs better and more human friendly.

Different interfaces need to be tested and worked with to find the best options.

Ted said...

Is it possible you could post the presentation? It seems that it doesn't display in swfdec and you have to create an account to download it.

Thank you,

Fitech said...

@Ted, I posted the original presentation.


mpt said...

Thank you for doing this study and for publishing the results. We are already working on improving most of the problem areas you uncovered. But it’s always useful to get specific test results to cite!

(At Canonical we’re interested in coordinating with groups doing user tests like this one. So if you’re reading this and planning a similar test, please get in touch.)

Matthew Paul Thomas
Canonical Design Team

Alex Railean said...

Thank you for sharing the results of your research.

I would like to point out that I was a bit worried by the light Windows bias of this report. For example, the fact that most Windows users expect the "change resolution" feature to be located in the right-click menu of the desktop doesn't mean that Linux should do it this way too.

Some things can be implemented in a way which is different from the Windows way and which also happens to be a better way.

Therefore I think that it is important for the survey to include "tabula rasa people" - i.e. they weren't exposed to Windows or any other OSes, so their feedback won't be a function of what they've seen in Windows.