No Section...

VII.2 March-April 2000
Page: 66
Digital Citation

Design brief: Stanford University


Authors:
Terry Winograd

About Stanford University

Stanford University in Stanford, California (http://www.stanford.edu) combines undergraduate and graduate education with renowned research programs in many disciplines, including computer science.

Philosophy of Education, Design

It is said in the real estate business that three factors determine the value of a property: location, location, and location. The joke is not so funny to people looking for affordable housing in our part of the country, but it is a good starting point for understanding what has shaped the HCI program at Stanford: its location in Silicon Valley and its location at a university with a tradition of interdisciplinary research.

The core of our philosophy is captured in our focus on "Interaction Design." By highlighting interaction as the object of design (rather than "interface" or "software" or "product") we emphasize the interplay of what people do with computer-based systems and what the systems do in return. By highlighting design as our activity (rather than "engineering" or "research" or "evaluation") we emphasize the kind of work that goes on in traditional design professions—the fitting together of technological possibilities with human needs and capabilities. Our curriculum is built around a philosophy of reflection in action, which integrates individual and group project work with coaching, reading, and discussion to provide a larger perspective.

Preparing Students for User Interface Careers

Our students come from diverse backgrounds. Some (such as those from medical informatics, product design, and the learning design and technology program) are headed toward careers in which they will apply interaction design techniques. Others, such as undergraduates in computer science and in the HCI concentration in symbolic systems (about 30 students each year) are headed into both industrial and academic careers. Several undergraduates have also developed individually designed majors in interaction design. The primary audience for many of our courses is the master of science program in computer science, with a designated concentration in HCI (about 20 students each year; see http://hci.stanford.edu/degrees/csms.html). Most of the MSCS HCI graduates work as designers, programmers, and managers in the computing industry. Advanced research students pursuing PhDs come from several departments, including computer science, communications, and medical informatics.

The introductory course, Introduction to Human-Computer Interaction Design (http://hci.stanford.edu/cs147) is based on case study and discussion rather than on project work, primarily because of its large enrollment (over 200 students). Students are often surprised that it appears to be based on unsuccessful products—examples we study include the Xerox Star, GO PenPoint, Microsoft Bob, and Action Technologies Coordinator. In each case, the product took some particular dimension of interaction to an extreme. Although this focus may have limited its commercial potential, it also brings out the unique issues in a clear way that fosters discussion of the design considerations and tradeoffs.

The introductory course is followed by a pair of project courses introducing the basic methodologies of interaction design. The Interaction Design Studio (http://hci.stanford.edu/cs247a) adopts a studio style from other design disciplines such as architecture, giving students a series of design exercises and critiquing the results. The course in Contextual and Organizational Issues in HCI (http://hci.stanford.edu/cs247b) takes a broader perspective on users and the worlds that they inhabit. Students learn and apply contextual techniques for user interviewing, observation, and work practice analysis. Small teams spend several weeks studying a site where new computer systems are being designed or adopted. Each team acts as a kind of mini-consulting group to its site, studying the contextual factors and identifying those that are critical to the design and/or success of the systems.

The rest of the curriculum shows the influence of the two key features of our location: the Silicon Valley environment and the Stanford environment.

First, there is a series of courses called Topics in Human-Computer Interaction (http://hci.stanford.edu/cs377). Each year, 3 to 10 courses are taught by volunteers from companies such as IBM, Xerox, Interval Research, Casio, Animatrix, Vivid, and Daimler-Chrysler. Topics have included Interactive Characters, Design of Domestic and Consumer Technologies, Computers as Persuasive Technologies, Filmcraft in User Interface Design, Adaptive Interfaces, Reactive and Proactive Computer Agents, Digital Multimedia Tools and Environments, The Design of Online Communities, Machine Perception for Human Computer Interface, and The Effects of Computer-based Communication on Media and Global Culture. These courses are generally (although not always) project based. A weekly speaker series, the Seminar on People, Computers, and Design (http://hci.stanford.edu/seminar), presents current research on a wide range of topics connected to interaction design. In the 10 years of this series, we have had over 250 speakers from industry and academia. The rich supply of teachers and speakers is one of the benefits we enjoy from our location in Silicon Valley.

Second, a number of departments offer courses related to HCI. A strong example is the work of Clifford Nass and Byron Reeves (communication department) on social responses to computer and communication technology. Most HCI students, from whatever departments, take some of their courses. Other options include courses on design methods and principles from the Design Division of Mechanical Engineering, courses on work practices and on entrepreneurship from the department of industrial engineering and engineering management, courses on medical applications from the department of medical informatics, and others including linguistics, psychology, music, education, and art. We also offer interdepartmental courses, such as the Interdisciplinary Interaction Design course offered jointly by mechanical engineering and computer science (discussed later here).

PhD students in HCI take some of these courses, but are primarily focused on research. Each student ends up pulling together a relatively unique combination of courses, advisors, and research activities. Stanford has a long tradition of encouraging research entrepreneurship—making it easy for faculty to create new entities for sponsored research. These range from large centers with dozens of industrial affiliates to individual laboratories consisting of a single faculty member with a few students. Research units cut across school and departmental boundaries, and many of them, such as the Center for the Study of Language and Information, are assertively interdisciplinary (CSLI brings together faculty from computer science, communications, linguistics, philosophy, and psychology).

This diversity provides research-oriented students with great opportunity and perhaps a little perplexity. There is no centralized research entity for HCI. A doctoral student in one department may end up affiliated with a research center housed in a different department with an advisor from yet a third. Dissertations and oral exam committees for HCI-related students almost always include faculty from more than one department. There is also the possibility of combining dissertation research with industrial research at one of the nearby laboratories, such as Xerox PARC, Interval Research, IBM Almaden Research Center, and SRI International. Given this diversity and the autonomy of each researcher and research unit, a student often needs to exercise some initiative and do some exploration to discover an appropriate research project. Once that is accomplished, there is a rich array of resources, both in people and other forms of support and collaboration.

One rapidly growing area of research collaboration at Stanford is with a new program for bioengineering, biomedicine and biosciences. The university recently received a major gift to support new initiatives in the intersection of biological sciences, medicine, and engineering, which will include a new building to bring together researchers. A number of current projects include medical applications of HCI, and we look forward to closer interactions as this venture progresses.

In order to facilitate communication and cooperation across the diversity of HCI-related work at Stanford, we have recently created the Stanford HCI Consortium with core faculty from a number of departments (http://hci.stanford.edu/faculty). This will serve as a forum to develop new HCI activities, both in teaching and research.

Design Project Example

A course on Interdisciplinary Interaction Design (http://hci.stanford.edu/cs447) is co-taught each year by Terry Winograd from computer science, and David Kelley from the product design program in mechanical engineering. It was inspired by the Apple Computer Interface Design Project, which was created by Joy Mountford at Apple in 1991 and continued until 1997. Mountford and her colleagues encouraged universities to initiate interdisciplinary design courses, in which student project groups competed for a place at an annual international student design workshop held at Apple. Apple provided initial equipment grants, mentoring visits from their researchers, and encouragement to go outside of the usual departmental boundaries. Stanford first participated in 1994, and the start from Apple led to an ongoing course.

The course is built around design projects by groups of students with complementary skills: from computer science, mechanical engineering, communications, business, and a wide scattering of other areas. Each project is guided through a series of milestone presentations reflecting the stages of our basic design methodology. Readings and lectures (many by guest lecturers from Silicon Valley industry) cover basic methods and skills, such as observation, interviewing, scenarios, storyboarding, character exemplars, prototyping, and usability testing. We place a great deal of emphasis on students finding a real user group and understanding their needs as a basis for designs. It is not possible for students to produce a full implementation given the short time span (the course is taught in a quarter, which leaves less than 10 weeks for the project, during which it is about one third of a student’s course load). Even in this short time, however, the teams produce polished prototypes, which reflect multiple iterations of discussion and testing with the prospective user group.

One group this year designed a communication tool for scuba diving groups. It features two different device interfaces, one for the dive leader (which includes a wrist display and a glove-based input device), and the other for the inexperienced members of the group, who are paired with "buddies" (Figure 1).

The team included students from computer science, symbolic systems, communications, and product design. Their research included talking with diving instructors (both before and after developing the prototypes), surveying related products, assessing the technical feasibility of underwater signal transmission, experimenting with buttons to be used under water with gloves on, user testing icons for clarity of meaning, and learning the standard hand-signal language for divers in order to design the glove input interactions. Only one of the students had previous diving experience, and much of the group learning was from discussions with potential users. In addition to the physical models illustrated in the figure, they produced a director mockup of the interaction sequences. This was included in a multimedia presentation for the end-of-quarter "show" to a panel of experienced interaction designers, which we held in an advanced presentation center volunteered by the local SAP Labs.

Although this project required a broad range of design activities—from shaping physical form and designing interaction to understanding the sociology and psychology of diving groups—the students were amazingly thorough in their thinking and in the professional quality of their presentations. Several of the projects over the years have been carried beyond the course, either within existing companies or in attempted startups (no billion-dollar successes yet). However, the learning value is not in the product as much as in the process—in providing an experience that integrates multiple disciplines and multiple design phases, letting students exercise their creativity while learning the methods and skills.

Author

Terry Winograd
Professor of Computer Science
Stanford University
Winograd@cs.stanford.edu
http://hci.stanford.edu

Figures

F1Figure 1. WetNet

UF1Figure. Terry Winograd, Professor of Computer Science, Stanford University

Sidebar: Departmental Snapshot

Degrees/Certificates Offered
Computer Science MSCS with specialization in HCI
Symbolic Systems BS with specialization in HCI
PhDs from various departments with dissertation topics in HCI (computer science, communications, medical informatics, etc.).

Sample Course Titles
Introduction to Human-Computer Interaction Design
Topics in Human-Computer Interaction
The Interaction Design Studio
Contextual and Organizational Issues in HCI

Sample Topics
Interactive Characters, Design of Domestic and Consumer Technologies, Computers as Persuasive Technologies, Filmcraft in User Interface Design, Adaptive Interfaces, Reactive and Proactive Computer Agents, Digital Multimedia Tools and Environments, The Design of Online Communities, Machine Perception for Human Computer Interface, and The Effects of Computer-Based Communication on Media and Global Culture

Sample Career Paths of HCI Program Graduates
Head of Interaction Design for Trilogy
Leader of Product Design for Financial Engines
Senior product manager for AOL Chat
HCI Researcher at Xerox PARC (several), NASA Ames
CEO of Internet startup companies (Google, E-groups)
Many are doing individual consulting or interaction design for Silicon Valley companies.

©2000 ACM  1072-5220/00/0200  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2000 ACM, Inc.

 

Post Comment


No Comments Found