Janet Davis

Systems directions for pervasive computing

Pervasive computing, with its focus on users and their tasks rather than on computing devices and technology, provides an attractive vision for the future of computing. But, while hardware and networking infrastructure to realize this vision are becoming a reality, precious few applications run in this infrastructure. We believe that this lack of applications stems largely from the fact that it is currently too hard to design, build, and deploy applications in the pervasive computing space. In this paper, we argue that existing approaches to distributed computing are flawed along three axes when applied to pervasive computing; we sketch out alternatives that are better suited for this space. First, application data and functionality need to be kept separate, so that they can evolve gracefully, in a global computing infrastructure. Second, applications need to be able to acquire any resource they need at any time, so that they can continuously provide their services in a highly dynamic environment. Third, pervasive computing requires a common system platform, allowing applications to be run across the range of devices and to be automatically distributed and installed.

Informing public deliberation: value sensitive design of indicators for a large-scale urban simulation

We investigate informing public deliberation regarding major land use and transportation decisions with the results from a sophisticated computer simulation of urban development. Our specific focus is on indicators that portray key results from the simulations. Our design addresses a number of challenges, including responding to the values and interests of diverse stakeholders, making documentation ready-to-hand, and balancing the value of fairness with presenting a diverse set of advocacy positions. We use Value Sensitive Design as our theory and design methodology; our theoretical framework also draws on Habermass theories of legitimation and communicative action. Our work contributes to CSCW as an example of designing a system for effective use in an environment with multiple stakeholders who have fundamental disagreements, and we conclude by drawing lessons for other environments with these characteristics.

A comparison of synchronous remote and local usability studies for an expert interface

Synchronous remote usability studies can be a convenient and cost-effective alternative to conventional local usability studies. Although they are common in the field, there has been little research comparing synchronous remote usability studies with local studies. In our comparison of remote and local studies of an expert interface, the primary differences were in the participants and facilitators qualitative experience. The remote and local studies agreed closely (with no significant differences) in terms of the number of usability issues found, their type, and their severity. While our comparison focuses on an expert interface and more work is needed to understand remote studies in general, our experience suggests that evaluators of expert interfaces will have comparable success identifying usability issues with either remote or local studies.

Design methods for ethical persuasive computing

Value Sensitive Design and Participatory Design are two methodological frameworks that account for ethical issues throughout the process of technology design. Through analysis and case studies, this paper argues that such methods should be applied to persuasive technology---computer systems that are intended to change behaviors and attitudes.

Simulations for Urban Planning: Designing for Human Values

Sophisticated simulation systems such as UrbanSim model the long-term impacts of transportation and land-use alternatives. Accounting for human values throughout the design process helps in designing interactions that engage both planners and citizens in the decision-making process. Our future work includes further research on applying UrbanSim in real planning contexts and on developing tools to support citizen discussion and comment. All these efforts are part of an overall agenda: to better sup port informed public deliberation and democratic engagement in the urban planning process

On edge-disjoint spanning trees in hypercubes

In this note we give a construction for obtaining the maximum number of edge-disjoint spanning trees in a hypercube. The result has applications to multicast communication in wormhole-routed parallel computers.

Food-first computer science: starting the first course right with PB&J

We consider in some depth a common exercise for the first session of a typical introductory computer science course: The task of writing instructions to make a peanut butter and jelly sandwich. The exercise, although simple, can engage students and motivate a variety of topics important throughout the semester. We discuss reasons to use such an exercise on the first day of class, present lessons students can learn from the exercise, and give practical advice for the instructor who wishes to make the most of this exercise.

Generating directions for persuasive technology design with the inspiration card workshop

Participatory design methods may help account for the ethical implications of persuasive technology. But how can participatory design methods both address ethical issues and lead to effective persuasive technologies? This paper presents the early stages of participatory design with a college EcoHouse. I discuss concepts resulting from an Inspiration Card Workshop [1], finally considering further development of participatory methods for designing persuasive technology.

ADAPT: audience design of ambient persuasive technology

We discuss our experiences with applying participatory design methods to the development of a persuasive ambient display. By combining these two approaches, we hoped to engage community members in reducing environmental waste on our college campus. We describe our design process and rationale, the resulting design, lessons learned, and future research directions.

Experiences with just-in-time teaching in systems and design courses

I report on my experiences with Just-in-Time Teaching (JiTT) in two systems courses and two design courses as a new instructor at a small liberal arts college. My purposes are threefold: to further raise awareness of JiTT in the computer science education community, to show instructors can achieve JiTTs goals with simple technology and preparation on the fly, and to consider strengths and limitations of the approach, particularly for new faculty. My experience has been very positive: classes have been fun, engaging, and conversational. Students have reported that they come to class better prepared and appreciate the opportunity to ask and discuss questions before class. Where others have warned of the high workload with this approach, my experience is that the workload is reasonable compared to preparing a new course from scratch; the approach discourages some types of overpreparation. For these reasons, I would recommend the approach for other new instructors.