Reach Out and Touch
College Park, Md.
Forget all the talk about the "on ramps" and "exits" needed to ease users onto the so-called "information highway." What's really missing is a nice, uncluttered dashboard and a few easy-to-read signposts.
So says Ben Shneiderman, the head of the Human-Computer Interaction Laboratory here at the University of Maryland's Center for Automation Research. Billions of dollars may eventually be spent to cable the proverbial "last mile" between the telecommunications industry's high-capacity transmission systems and individual users in the nation's homes, schools, and libraries. But if those wires dead-end in computer keyboards, Shneiderman argues, the number of people willing to go along for the ride will be drastically limited.
"The last mile is a problem, but it's somebody else's problem," Shneiderman admits. "To me, the problem is the 'last yard,' what happens from the screen to the user's head."
Bearded and elfin, the 47-year-old Shneiderman is one of the world's leading experts in the design of computer interfaces. He's paid, in short, to think about how to make it easier and more intuitive to retrieve, store, process, and share information by computer.
The most common interface in use today, for example, is the computer keyboard. It's often coupled with the "mouse," an invention dating back to the mid-1980's, the dim past for a field where innovation is the key to survival.
Shneiderman is willing to concede that the mouse is a relatively easy-to-learn and effective navigation tool. But he doesn't have much good to say about keyboards. In fact, he suggests throwing them out.
Instead, Shneiderman advocates more user-friendly touch-screens, touch-pads, and other "visually oriented pointing mechanisms" that minimize the physical barriers to interacting with a computer.
He points to a touch-screen system developed in 1992 by researchers here in conjunction with the Library of Congress as one more intimate interface. ACCESS, an on-screen menu of options, allows patrons--even inexperienced ones--to search the library's boundless stacks simply by touching the appropriate spots on a video screen.
Two years later, the ACCESS system has largely superseded the library's existing command-line interface, known as SCORPIO, which requires four to six hours of training to use effectively. Although the more powerful SCORPIO interface is still available to professional researchers visiting the library, ACCESS has become the on-site interface of choice for the general public.
Giving every student electronic ACCESS to all of the information housed in the Library of Congress has become almost a mantra for those heralding the educational benefits of building a truly ubiquitous information highway.
Vice President Gore, for example, "is wonderful as a promoter of this scenario," Shneiderman says. "But somehow, we've got to get in there and do something to make it possible."
"Right now," he adds, "if students dial up the Library of Congress, all they're going to find is a card catalogue and the old SCORPIO interface."
Until recently, how people interacted with their computers was of interest only to a handful of researchers, software engineers, and computer manufacturers.
Granted, the phrase "user interface" has a disturbingly impersonal ring. But, Shneiderman says, an increasing number of lay people have started to realize that interfaces influence how successfully users integrate a given technology into their daily lives.
Consider, for example, the difficulty many people have programming their videocassette recorders. In fact, the elusive function has made VCR technology one of the most common examples of what can happen when an industry fails to design a user-friendly interface.
User-friendliness will be particularly important to potential classroom users, observers note, because, as a group, educators have been hesitant to embrace technology and are seldom given incentives to do so. Given the minimal time and resources school districts generally devote to training, an easy-to-use interface is vital to bringing more teachers and students onto the information highway.
Despite stepped-up efforts to make technology more ACCESSible, Shneiderman warns that the interface issue could become a major obstacle to the Clinton Administration's hopes for a universally ACCESSible "National Information Infrastructure."
Although wary of heavy-handed government intervention in developing the information highway, Shneiderman suggests that just as the Federal Highway Administration sets standards for road signs, a federal "Information Highway Administration" must map out the digital terrain. Such an agency would "act as an agent of structure and consistency" in the digital age, insuring that electronic interfaces conform to straightforward usability standards.
"The government should say to whoever builds these highways out there: 'These are the colors for the dotted lines, these are the ways that the exit ramps are marked, and this is what a stop sign looks like,'" Shneiderman argues. But so far, he says, the Administration has failed to emphasize the importance of the interface question in its thinking about how to make its National Information Infrastructure a valuable public asset.
"'Easy to use' is easy to say," Shneiderman says. "Building it requires a lot of effort."