Usability and Mobile Devices
By Daniel Roberts
Topics on this Page:
- Unique Problems
- Impact of Low Resolution
- Impact of Download Speed
- Impact of Limited Screen Size
- Screen Readers for Mobile Devices
- RSS Feeds
- Physically Impaired
- About the Author
As of May 2008, there were 40.04 million unique users accessing the mobile Internet at least once a month (Nielsen, 2008). This number increased by 73% over a period of two years (i.e., 2006-2008). As the prevalence of PDA and other smart devices with Internet connectivity increase, so does the need for attention to the usability of these devices.
Much of the body of Web usability literature has been done on computing in a stationary environment. And although much of the existing research applies to the mobile Web, some principles need to be altered to address the unique problems associated with this form of Web access.
Mobile devices present several unique problems in addition to the standard usability issues that challenge Web designers. The issues associated with mobile devices are a function of the devices (Zhang & Adipat, 2005): small, low resolution screens; limited input option (e.g., no mouse or full keyboard); slow computer hardware; and slow or sometimes unreliable connectivity.
These problems are exasperated because despite some Websites with alternate mobile designs, most do not. Most sites are designed for full scale PCs or laptops and they do not take the mobile user into account (Borodin et al, 2005).
Jones, et al (1999) investigated the effects of using a low resolution screen to view Web sites. Participants were assigned to one of two groups either working on a desktop display of either 1074 X 768 or 640 x 480. Participants had 15 minutes to complete four information search tasks. Participants using the high resolution screens were able to answer twice the number of search tasks correctly and spent less time scrolling then their low resolution counterparts.
Compared to desktop Web users, mobile users spend more time attempting to locate information rather than simply browsing. As a result of their findings, Jones, et al had these recommendations for mobile interfaces:
- Minimize the amount of scrolling required to find information
- Include more search options
Roto and Oulasvitra (2005) looked at the impact of the usability issues that surround content download speed in a mobile environment. Of driving concern was that mobile devices are often designed without concern for the environments in which they are used. These environments are sometimes uncontrolled and involve more distractions and competing tasks than may exist in the typical desktop environment.
Roto and Oulasvitra had participants wear video cameras while simultaneously traversing a city and performing a Web task on a smart phone. A comparison group performed the same mobile Web task while seated in a laboratory setting. Results showed a difference in average gaze time before looking away was 14.3 seconds in the laboratory compared to an average between 4 to 8 seconds for mobile users.
Roto and Oulasvitra suggest implementing a tactile feedback mechanism to notify users if the task is expected to take more than four seconds. The feedback would allow users to direct their attention back to the environment rather frequently checking the screen status. This should reduce the amount of needed gazes and increase user safety.
One of the most salient usability issues for the mobile Web is the limited screen size. Rather than having to develop a discrete mobile version, researchers have been developing methods to make existing versions more mobile Web-friendly.
These same structural strategies will also benefit desktop Web users with reduced vision (Horton, 2006). Both mobile and sight-impaired users lack the ability to scan an entire page to locate information. Horton’s changes include:
- Separate HTML mark-up language from content information using CSS
- Use a consistent, uncluttered page design
- Limit the amount of text used for non-semantic purposes
- Arrange information in an inverted pyramid styles (from most to least important)
Several researchers have introduced screen reader equivalents for mobile users. The software converts normal Web pages into a mobile friendly format. Borodin, et al (2007) described software (CMo) that analyzes the semantic content of each requested page and displays it in order of predicted importance to the user question.
The accuracy of the CMo algorithm compared to human raters was high. The system identified the correct context on the linking page 80%-90% of the time and the correct block on the linked page an average 91% of the time. When compared to typical users, the CMo system versus a traditional PDA browser improved user performance by 46% in time taken to complete the task, and 41% in terms of the number of stylus taps.
Garofalakis and Stefanis (2007) described a system that takes advantage of the existing formatting of content into RSS feeds. The authors used an algorithm that takes content formatted as RSS and adapts it to the mobile screen.
Garofalakis and Stefanis asked participants to locate a Web news article within three minutes and compared their performance to the existing mobile browser. Results showed improved success rates on the search task with the new system as well as positive useful ratings by participants.
Chen (2008) explored the comparison of mobile users and physically impaired users but in the context of input rather than output. Chen identified the types of errors by motor impaired Web users. He asked non-physically impaired users to perform the same types of tasks but using a mobile Web device.
Results showed that non-physically impaired users have the same error rates as physically impaired on tasks that involve typing or pointing. Chen suggests that the existing body of accessibility literature and technology may be applied to mobile Web users as well.
The mobile Web poses several unique challenges to usability researchers and designers. These challenges are compounded by the mobile Web environment being only a part of the Web universe. Improving functionality may require adapting Web content to the specific needs of the mobile user rather than the current condition where all content is designed for the Web.
What may be most needed in mobile/Web usability is a standardization of the usability process so comparisons can be easily made between devices and systems. Fortunately this approach has been adapted by Zhang & Adipat (2005). They describe a framework for researchers to identify the best methodologies for answering Web usability research questions.
Borodin, Y., Mahmud, J., & Ramakrishnan, I.V. (2007). Context browsing with mobiles – When less is more. In Proceedings of the 5th international conference on Mobile systems, applications and services, 11-17, June 2007 (pp. 3-15). San Juan, Puerto Rico: ACM.
Chen, T. (2008). Input to the mobile web in situationally-impaired. In Proceedings of the 10th international ACM SIGACCESS conference on Computers and accessibility. 13-15, October 2008 (pp.303-304). Halifax, Nova Scotia, Canada: ACM
Garofalakis, J., & Stefanis, V., (2007). Using RSS feeds for effective mobile web browsing. Universal Access in the Information Society, 6(3), 249-257.
Horton, S. (2006). Designing beneath the surface of the web. In Proceedings of the 2006 international cross-disciplinary workshop on Web accessibility, 23-26, May 2006 (pp. 1-5). Edinburgh, U.K.: ACM
Jones, M., Marsdens, G., Mohd-Nsir, N., Boone, K., & Buchanan, G., (1999). Improving web interaction on small displays. Computer Networks, 31, 1129-1137.
Neilsen Mobile (2008) Critical Mass: The Worldwide State of the Mobile Web. [Electronic version]. Retrieved November 2008 from http:www/nielsenmobile.com/documents/CritcalMass.pdf
Roto, V., & Oulasvitra, A. (2005). Need for non-visual feedback with long response times in mobile HCI. In Special interest tracks and posters of the 14th international conference on World Wide Web, 10-14 May 2005 (pp. 775-781). Chiba, Japan: ACM
Zhang, D., & Adipat, B. (2005). Challenges, methodologies, and issues in the usability of testing mobile applications. International Journal of Human-Computer Interaction, 18(3), 293-308.
Daniel Roberts is a second year Master’s student in the George Mason University Human Factors and Applied Cognition program. His research interests include auditory cognition and mental workload assessment. He wrote this research paper as a course requirement.