Design & Evaluation of Context-Aware Computing

Context-Aware Computing

A context-aware system is one that reacts to its environment. Such a system exhibits a level of intelligence by presenting information based on user behavior, such as where the user is physically located. The system’s reaction to the user in turn elicits a wide range of user reactions that are often not anticipated. Context-aware technology is becoming more common, but little has been done to evaluate user response for incorporation into future designs.

Iterative Design

The iterative design development model is an effective way to develop context-aware systems. This model relies on using user feedback from a completed design to develop a new design. Iterative design is a cyclical process, moving from Requirements, to Design, Implementation, Evaluation, and then starting another cycle to make a new version.

The Cornell Human-Computer Interaction Group uses this model in all of its projects. After a technology prototype is created, it is evaluated, recreated, and evaluated again to find better and more refined approaches to solving a problem.

HCI Projects

The HCI Group has been exploring context-aware systems through the CampusAware and MUSE projects. These are ongoing projects, both having multiple revisions, that center on how handheld devices can be used to improve a visiting experience. With each revision, the systems were evaluated and redesigned to incorporate the successes of previous designs and improve shortcomings.

CampusAware

CampusAware is a college campus tour guide that provides relevant information based on a user’s location. What makes this system different from traditional location-aware systems is that much of the content in the system is created by users. A user can leave a note in the system at a particular location for future users to see. CampusAware allows the user to move at his or her own pace and share thoughts with other visitors.

CampusAware evolved from a previous system called E-Graffiti. This system was used by full-time students on wireless laptops, usually in class or a library, to leave virtual notes at locations around campus. CampusAware is an overhaul of E-Graffiti applied in a different scenario. The system relies on Palm Pilot PDAs instead of laptops and uses GPS to provide more accurate location detection.

MUSE

MUSE is an application written for PDAs to enhance a museum experience with navigation tools and encourage social interaction with other users or museum personnel. In addition, users are able to extend the museum experience beyond the walls of the museum with access to all the information in the system using a standard browser.

A descendent of CampusAware, MUSE uses beacon scanning at each exhibit rather than a GPS system. The system has instant access to text, images, and audio streams. Like CampusAware, users can post questions and comments and leave them at the exhibits for other visitors.

BOOM 2003

At the Bits On Our Minds 2003 fair, we will be showcasing the latest versions of CampusAware and MUSE. The presentation will highlight design issues, the evaluation process, and results from user testing.