obso1337.org

Commentary: This report does not include the survey comparison to the work completed by the SUSE Interfaces and Applications Team. I ran out of time (and writing space, the paper is twice longer than required) and will provide a separate analysis sometime in the future.

Abstract

“Start menus” are a popular desktop menu system meant to easily provide application shortcuts to users. Although the original context of these menus worked very well, the menu system has not evolved with user’s information habits. Usability issues with the menu system are known, however design changes to existing menus and the design alternatives of replacement menus do
not seem to fulfill user’s information needs.

In this study, three areas of usability were examined: user research and requirements,
information architecture, and interaction design. It was clear from the user survey that participants had a certain level of dissatisfaction for start menu system; they preferred methods other than their start menu to launch applications, and expected functionality which for the most part does not currently exist in the menu systems. Information architecture was an extreme issue
for the systems analyzed, both on a fundamental structure and detail quality level. Task analysis and interaction for two menus was simple and efficient, without noticeable issues with cascading horizontal menus. Task analysis and interaction for the larger menu was extremely complex, however some of the menu features provided alternatives which were more efficient than the
other two menus.

There are serious issues with start menu systems, particularly the suitability for addressing users needs and its information architecture. Certain properties of the information architecture can be addressed in the short term and will greatly improve item findability. However additional research must be conducted on user’s information needs and information architecture concepts
before a long term solution can be designed.

Read the rest of the paper: (PDF 632KB)

seele :: Aug.29.2006 :: Academics, General, KDE/Kubuntu, Usability :: 9 Comments »

on 29 Aug 2006 at 5:02 pm1Marc Cramdal

Hello !

Great job ! It’s really interesting as the work on this kind of menu seems to begin.

Have you heard of Katapult. I find it a very usable way to launch programs (quicker than minicli I think)

Marc.

on 29 Aug 2006 at 6:56 pm2Dark Phoenix

After reading through the report, I suddenly thought of a question that I had once wondered at before.

With Linux’s ability to use links to make files appear to be in multiple places at once, why is it that both the KMenu and the GNOME Applications Menu require duplicate entries if you want programs in more than one category? Considering the survey notes how annoying people find it to look in a certain category and not find a program that should be there, one would think this would be used to allow multiple entries that would be updated in sync…

on 30 Aug 2006 at 12:44 pm3RJames

Concerning menu links:
isn’t the kde & gnome menu system handled through an xml schema?

http://www.freedesktop.org/wiki/Standards/menu-spec
And who says freedesktop org doesn’t handle standards?

on 31 Aug 2006 at 5:09 am4jhs

An error is on page 2, the last line of the “Methods” section. The correct word is “affect,” not “effect.”

on 31 Aug 2006 at 9:07 am5numerodix

I actually had no idea that people were doing this kind of comprehensive studies on UI design, applying statistical metrics and whatnot just to analyze the merits of one particular design idea, not in open source anyway. While that probably makes me sound incredibly ignorant, I think it’s a great thing that it’s being looked at in-depth to this degree, and not only by the ever so clever people at Redmond.

Power to KDE! :)

on 01 Sep 2006 at 9:20 am6Bobulate

Events and sorts of “hello”…

Ahh, September. Isn’t it labour day for some around now? Labor perhaps. One of my research projects has British English as its official spelling while KDE uses en_US and I naturally write en_CA which is a weird mix of the two. In any case we made it t…

on 01 Sep 2006 at 4:06 pm7wspeir

Haven’t finished reading the study but one thing really caught my eye:

In ‘How much importance do you attach to…’
“The most important feature was the acoustic feedback of clicks (4.0/5.0). This seemed very strange…”

I assume this is the OS playing a sound back through the speakers when the mouse is clicked.

This drives me absolutely batty! Every mouse I’ve ever used gives me a subtle acoustic feedback already along with tactile feedback. I know when the mouse button has been clicked from this feedback already.

To add another sound generated by the OS completely disorients me. It feels out of place. It’s pratically visceral.

Wow, I’m getting carried away. I guess I didn’t realize how much I hate it.

Anyhow, it’s the first thing that gets turned off on any computer that I use.

Thanks for your work…

on 01 Sep 2006 at 4:08 pm8wspeir

I made sure to spell visceral correctly but somehow missed “practically”…

on 18 Sep 2006 at 8:27 am9Hebi to panda no blog

Els noms dels programes lliures són massa frikis…

L’altre dia vaig llegir en una notícia1 que algú deia que Linux no estava preparat per als usuaris "normals", ja que se’ls feia difícil buscar el programa necessari per fer una cosa determinada. Que els desenvolupadors s’havien d’o…

The adventures of Human-Computer Interaction in Open Source Software

Study of Desktop “Start Menu” System Usability

9 Responses to “Study of Desktop “Start Menu” System Usability”

Recent Posts

Recent Comments

Upcoming Events