Articles

---

Ben Shneiderman on how designers can help people succeed

Ben Shneiderman is a Professor in the Department of Computer Science, Founding Director of the Human-Computer Interaction Laboratory, and member of the Institute for Advanced Computer Studies and the Institute for Systems Research, all at the University of Maryland at College Park. He is a Fellow of the ACM, a Fellow of the American Association for the Advancement of Science, and founding chair of the ACM Conference on Universal Usability.. In 2001, he received the ACM Computer Human Interaction Lifetime Achievement Award. In this interview, Shneiderman talks about his new book, "Leonardo's Laptop: Human Needs and the New Computing Technology".

UBIQUITY: What do you mean by the title of your new book, "Leonardo's Laptop: Human Needs and the New Computing Technology".

SHNEIDERMAN: Many users of today's computer systems suffer and experience frustration in trying to use their computers. I think every designer would like to envision a future that is graceful, elegant, comfortable, satisfying, comprehensible, predictable and controllable. Those are the adjectives that we'd like to attach to the interfaces and systems that we use. The book calls for a new renaissance and fresh thinking about user needs -- elevating the expectations of users and the responsibilities of designers. This focus on users leads me to the book's repeated refrain: The old computing is about what computers can do; the new computing is about what people can do. Designers can do better in helping users succeed. Too often people are struggling because they can't understand the menu choices, they don't know what the dialog boxes mean, and the error messages are too frustrating and confusing. Attachments won't open. Viruses intrude on their experience. Spam clutters their e-mail inboxes. I have found Leonardo da Vinci to be an appropriate inspirational muse for the new computing. He combined art and science and aesthetics and engineering. That kind of unity is needed once again.

UBIQUITY: Have studies been done to show how pervasive these problems are?

SHNEIDERMAN: One study suggests that users waste 5.1 hours per week trying to get their computers to work. Our recent study of user frustration found that it's even worse than that -- approximately 45 percent of the users' time was spent in dealing with computer-related problems.

UBIQUITY: You certainly won't have any trouble convincing people that they waste a lot of time and that they are frustrated. Will you have trouble convincing people that anything can be done about it?

SHNEIDERMAN: As we do these studies, we're finding that there are many problems. There is no silver bullet that's going to solve them all. We need to find a systematic way for designers to develop and test software to make it more reliable. Another way is to build in the proper motivations or penalties for developers; one proposal is that every time an application crashes your computer files a report electronically and you get $1 credit from the developer. They get data about failures, and you get a small reward/compensation.

UBIQUITY: Is there any evidence that software companies would actually support this idea?

SHNEIDERMAN: Well, both Netscape and Microsoft have features that report errors. You don't get money, but they do file a report by e-mail. Airlines compensate you for delayed luggage, so why shouldn't you get something when software goes bad? One group I'd like to give a plug for is a company called www.bugtoaster.com. Their Web site reports on 280,000 crashes indicating which operating systems and applications crash most frequently. As a community, we need to capture the data, and then create the incentives for developers, maintainers and manufacturers to improve products so that users have a better experience dealing with computers.

UBIQUITY: The first chapter of the book is a play on the title of Ralph Nader's 1965 book "Unsafe at Any Speed". What do you mean by "Unusable at Any Bandwidth"?

SHNEIDERMAN: Just going to broadband will not solve the users' problems. It's a matter of improving the quality of the user experience. Then we need to enable a broad community of users to participate in the system. We have to address the goal of universal usability: helping people whose English skills are poor, people working on slow modems, small displays and old computers, people whose reading may be poor, or whose motivation may be low. We think there can be dramatic improvements to systems and enormous benefits to users.

UBIQUITY: What areas will be most improved by these types of innovations?

SHNEIDERMAN: The four application domains that have the biggest opportunities to improve the quality of life of users are e-commerce, e-learning, e-government and e-healthcare. In each chapter, I talk about how Leonardo addressed these issues in his time. Then I take that message and interpret how we might use it to create more satisfying experiences today. In e-healthcare, for example, the first idea is the World Wide Med: all your medical information would be available wherever you were brought to an emergency room. Within 15 seconds, your medical records could be on the screen for physicians and healthcare professionals in their local language. Of course, protection of privacy is essential in this environment, but the potential to make that information available is an exciting challenge that will bring benefits in many ways. It will improve the care of patients, but also clinical research by understanding of the spread of epidemics, prevalence of diseases, and efficacy of treatments.

UBIQUITY: How does that chapter tie in with Leonardo?

SHNEIDERMAN: Leonardo was a brilliant student of anatomy and we could imagine that he would promote medical image databases. He advanced the state of knowledge of medicine by his detailed drawings of the human body. He participated in about 30 autopsies, a controversial practice in his day. The style of his medical drawings -- cutaways and three-dimensional presentations -- is still an inspiration to medical illustrators. He almost got to the principles of the circulation of blood through the heart and lungs, but it took another 150 years till William Harvey made the leap to that understanding. Leonardo was a superb observer, and he always asked himself questions about what he was seeing.

UBIQUITY: Your book talks about another development in healthcare, the super patient. Do super patients have superior knowledge?

SHNEIDERMAN: Today, some patients arrive at the physician's office knowing more than the physician about their disease and experimental trials that are currently underway. They become stronger partners in their medical care and can share the responsibility for decision-making. The availability of the Web and online discussion groups has enabled more people to know more about their medical care. This challenges some physicians who are uncomfortable with probing questions from their patients, but most doctors are pleased to find their patients knowledgeable and ready to participate actively in their medical care.

UBIQUITY: What are some of the potential transformations in e-government?

SHNEIDERMAN: Million-person communities might change the consensus-seeking process. It won't be a million people just doing e-mail or listservs. We will evolve group discussions in which participants build consensus through rational discourse to build support for taking action. The potential to create more active discussions with greater trust and empathy is very attractive. You can envision discussions among small groups, proposals being put forward and building up towards having the whole group make a decision. I try to outline how group processes could occur within the new computing. There's another scenario in the book about an individual who wants a stop sign placed in front of his neighborhood school. In the new computing, he could collect information, have satellite photos, compare the traffic rates at this intersection and other intersections, seek the support of his neighbors, and prepare a report in the standard format to the appropriate agency, thus creating a community decision by having access to more information.

UBIQUITY: Another change that you and many see is in technology and education. What does your book add to the e-learning discussion?

SHNEIDERMAN: The e-learning section talks about how collaborative projects and active learning might be developed. The strategy I call 'relate-create-donate' suggests that students work in teams to produce something ambitious for the benefit of someone outside the classroom. The collaborative environment of the Web means that it's no longer acceptable that students write papers that are read only by professors and then thrown out. My students' projects for the last eight years are on the Web at www.otal.umd.edu/SHORE2001. Students work in teams to conduct experiments on user interface issues. One early, simple project on the size of screens showed that if you had larger screens, users got their jobs done faster. The study was cited by manufacturers of large screens. It was a research result that produced a worthwhile research contribution.

UBIQUITY: How has this idea of producing work for the benefit of someone outside the classroom affected the quality of your students' work?

SHNEIDERMAN: I used to say that one out of ten student projects had something worth publishing. At this point, I would say three or four out of ten have something that's worth publishing. The capacity of students to do high-quality work has dramatically improved. Furthermore, the dropout rate in these courses is low. You see every student's project on the Web. Early on the students ask the question, what if my project's not good enough to be on the Web? The question I bounce back is, how can we work as a class to make sure everybody's is good enough? So the students read and comment on each other's projects and make recommendations for improvements. Then they have another period to make improvements. Over the years, I've evolved to having 15 percent of students' course grades determined by how much they improve their colleagues' work.

UBIQUITY: Going back to the Leonardo theme, how does your book address the issue of user frustration?

SHNEIDERMAN: The subtitle of the book suggests that human needs should be the guide for our new technologies. So the intellectual heart of the book says, when we look for new technologies and ways to guide the future, then we, like Leonardo, should think about humans and their needs. Leonardo was concerned about commerce, security, education, medicine and urban design. He was very eclectic in his thoughts and he was also very practical. He wanted to see real things get built. He also was concerned about aesthetics. The intellectual heart of the book is a new understanding of human needs defined by the relationships people have and the activities they carry out. Instead of trying to apply the latest new technology, we start by asking what is it that people want to do in their lives? This theory starts by recognizing that there are some things you do on your own. There are other things you do with family and friends, others with neighbors and colleagues, and yet others with citizens and markets. So the first understanding is how we would change applications that are built for personal use into ones that are for a small group of intimate trusted friends and neighbors, people who you have a shared knowledge with, who you will see again, and who wouldn't harm you. Things like security are less important for that community. As you go out to larger circles, the neighbors and colleagues, and citizens and markets, issues of security and shared understandings become ever more important. The second dimension of this theory is human activities.

UBIQUITY: How do you define the activities?

SHNEIDERMAN: I separate the activities into four groups. The first is collecting information -- the information superhighway concept. Next is forming relationships with people through communications media -- e-mail and instant messaging and so on. The third category is making innovations -- the creative works that you do, whether in science or art or commerce. The last category is disseminating. We need better tools for disseminating or donating to others in order to bring the results of our innovation to a wider range of appropriate people. This four-by-four matrix that forms the Activities and Relationship Table (appropriately abbreviated as ART) is the heart of the book. Then the book applies this table to the four domains of e-learning, e-healthcare, e-government and e-business.

UBIQUITY: Give a real-life example of how you apply your theory.

SHNEIDERMAN: One of the examples we've worked on is photography systems. We worked with the Library of Congress to build a photographic information system where citizens could get Mathew Brady's photos of the Civil War or architectural photos of Washington. That was wonderful, but it was just collecting information. Users couldn't add anything. They couldn't do much with the photos they viewed. In recent years, we've built a personal system, called Photo Finder, for doing family photos. As we built that system, we faced a very different set of constraints. We wound up with a quite exciting and novel product that you can see on our Website: www.cs.umd.edu/hcil/photolib. We've moved that from a personal photo library to a community photo archive, annotated by hundreds of people. We've just put up a photo history of the ACM SIGCHI group: www.acm.org/sigchi/photohistory. As we redesigned our photo system to support a community of people, we had to change our design dramatically. The article about it appears in the May/June 2002 issue of the ACM Interactions. Redoing photo systems from personal to public and allowing people to create something from their photos opens new possibilities. Users can make greeting cards, record family histories, tell stories, and start new businesses.

UBIQUITY: Is your book a call to computer scientists? Is that the target audience?

SHNEIDERMAN: There are two audiences. The primary audience is the general reader, and the hope is to raise their expectations of what can be gained from using computers. I hope my ideas invigorate discussions of e-commerce, e-learning, e-government and e-healthcare. The second audience are the professional designers who, having considered user needs for activities and relationships, will have a fresh perspective on creating innovative products. Successful innovations will be usable, universal, and useful, thereby empowering people to improve the quality of their lives.