In the spring semester 1997, Ben Shneiderman taught an undergraduate course on human-computer interaction at the University of Maryland, College Park. Dr. Shneiderman, the Director of the Human-Computer Interaction Laboratory, is a well-known researcher and a founder of human-computer interactions (HCI) and the social informatics movement, which D-Lib will cover more thoroughly in the coming months. The third edition of his book, Designing the User Interface, has just been published. HCI gets at some fundamental questions in computing, cognition, and searching in which the web interface is the mediating level. He graciously agreed to talk to us.
Amy Friedlander
Editor
The world has seen technological innovations before; what's different now? And what has been brought in when the information technologies enter the mix?
The work place that our students will enter and our educational institutions are dramatically changing as information technology becomes widespread. For me, there are four parts: First, there's the empowerment of the individual by better access to the existing literature. It's amazing and transformational.
Second, we achieve this empowerment through creative tools from word processing to spreadsheets to simulation and model-building tools. We can build faster and push them further. At Maryland, we have a project called SIMPLE (Simulated Processes in a Learning Environment), which investigates how to scaffold these processes to make them educationally more productive; these educational tools are simply remarkable.
Third, we now have the capacity for consultation and cooperative work enabled by groupware and community networks. Students can collaborate with peers and obtain input from mentors. I expect my students to be in touch by e-mail with the authors of the papers that they have read. The authors are usually thrilled to know that somebody cares about their work. The response has been wonderful. And students have become acknowledged in subsequent papers. I get a lot of e-mail myself from unknown students around the world.
Finally and related to these interactions, once they have obtained feedback from peers and mentors, the students are expected to publish and disseminate their results, to communicate with communities that are relevant. Pushing them to become part of the community of scientists improves their work: through access to the literature, peer review among themselves, and the pressure that results from the knowledge that their work will be accessible on-line. Since the workplace demands collaboration, I think that my responsibility as teacher is to train my students and give them experience in many modes of teamwork.
So they are partly behaving like scientists and partly learning to collaborate?
Yes, they have to plan the work, set schedules, critique each others' work, and integrate components to produce a polished, finished product.
We've talked about you as a teacher; now let's talk a little about you as scientist. What are you working on now?
Our earlier work on hypertext led to the "hotspot", which has become widely applied in the World Wide Web. The idea of clicking on words in a sentence was developed here in 1982-1985 in our work for the United States Holocaust Memorial Museum. Links had previously been by specifying codes, numbered menus, anchor icons, and so on; the idea of making the text clickable originated in the work here. And it has permeated the web. It is one of those wonderful things; I wish I had patented it.
Now our emphasis is on information visualization, especially for digital libraries. We've spent the last two years on the National Digital Library Program for the Library of Congress, developing advanced prototypes for American Memory. Bits and pieces have found their way in their implementations, and we've delivered a set of specifications to the Library, which show them ways to browse libraries of photographs and provide visual overviews. WebTOC is available as a free download. It goes out to anybody's site and produces a hierarchy created out of the web site, which provides a visualization of the amount of the stuff that's there and where differently formatted materials are. So if you're looking for videos, for example, you will see where they are in a very large web site instantly.
Visualization is a hot direction right now. One of the things we've seen at Maryland is a new spin-off company based on our visualization work. A former student Chris Ahlberg has formed a company, Spotfire, which provides generic information visualization tools that can be used for digital library applications.
What about digital libraries and their possible relationships with other movements, like social informatics?
Well, one thing I'd like to see you do is change the title from the "Digital Libraries Initiatives" or "D-Lib" to the "Digital Libraries Communities Initiative". Ignoring community aspects is a fundamental mistake: in its best moments, a digital library is a community. There is community of librarians and of users; this stuff doesn't just happen magically.
It's troubling to me that many people do not distinguish between people and machines. Here, Doug Engelbart is a hero; he got the idea early that computing was about human augmentation. And Vannevar Bush [Franklin D. Roosevelt's science advisor, who first envisaged hypertext via something he called the "Memex" in 1945] really had the idea first - putting the human at the center. I tried to think what would come after Memex and came up with Genex - meaning "generator of excellence," which has the four components I described earlier: collecting, creating, consulting, and contributing.
Bush was revolutionary partly because he put the user at the center; he imagined what is essentially a workstation, admittedly with microfilm, but a setting in which the information surrounded the user, who was equipped with tools to work with it.
Yes, but the pendulum has swung back again - from the personal workstation back to the web - where information is spread out in a dynamic environment but brought back to the user. But it is important to make the statement: designers of digital libraries must think about people; digital libraries will fail unless the designers understand how to create communities of shared interest. Of course, the communities are ever-changing. How wonderful.
That's my major criticism of the current Digital Libraries Initiative; they haven't spent enough time on users. I think they should shift the emphasis by retitling it to the Digital Library Communities Initiative.
The goal is insight not numbers. The purpose of information is insight not disc drives or databases. Students and other users of digital libraries can be inspired and thrilled by the power they have to be creative. The purpose is to build a better world, to make it wiser, warmer, safer, and more joyous.
September 5, 1997
hdl:cnri.dlib/september97-editorial