YOU ARE HERE: LAT HomeCollections

Digital Nation

To Be More Usable, Desktops Need to Take a Page From Hand-Helds

February 15, 2001|GARY CHAPMAN |

Here in 2001, we have millions of people sitting in front of awesomely powerful desktop computers. A lot of these people are either ignorant of their software's absurdly bloated features or they couldn't care less about the features they don't need or use. And most are routinely exasperated by the ways these machines work or fail to work.

What's wrong with this picture?

One reason this question comes up so often is the stunning success of the Palm and its spinoff and competitor, the Handspring Visor. In the midst of a plunge in PC sales, hand-held personal digital assistants have doubled in sales in the last year, and people seem to love them, especially the easy-to-use devices based on Palm's operating system.

The Palm was designed to do some very simple things well, and its interface is so straightforward and intuitive that desktop and laptop computers now suffer greatly by comparison. The Palm is instantly on, it typically presents everything it can do on its first screen, it's nearly impossible to lose data in it as long as its memory is powered, and, of course, it's conveniently portable and efficient with its batteries.

The Palm is much more than just a successful gizmo--it may be pushing computing in some interesting design directions. Slowly, a new paradigm for personal computing seems to be emerging.

Granted, the Palm's small screen is limited to certain tasks; consumers wouldn't want to prepare documents on it, and fast data entry is difficult. But many have adjusted to these limits by adding a keyboard or a modem.

"The old model of computing was about what computers can do," said Ben Shneiderman, a professor of computer science at the University of Maryland and one of the leading thinkers on usability and interfaces. "The new model of computing is about what users do.

"It's certainly time to get angry about the quality of [PC] interfaces," Shneiderman said. "There are just too many frustrations in everyday use, and the public is tired of the industry's excuses."

Among other changes, it seems very likely that computers of the future will be always on, always connected to the Internet and probably getting information from the network even while the user is doing something else. They should be rock-solid stable. Files and programs will always be available, no matter what computer is being used and where it is.

User interfaces and software will serve people with a variety of skills and needs, even people who use the same computer. Consumer access to files and information and personal computing preferences will follow you around, rather than staying stuck on a single machine. You'll probably use different devices for different purposes, or in different locales.

Shneiderman also has proposed a feature for all software that would let users choose what level of expertise, or how many features, their operating system and applications will support.

A step in this direction is reflected in an operating system called Athena, from Rocklyte Systems ( The Web site says that the new approach is a "multi-platform operating system that provides users with a totally configurable, XML-based graphical interface." Athena can look like any familiar operating system or something completely novel, and it can be simple or complex according to user preference. Currently, it's available for Linux computers, but Rocklyte promises versions for Windows and Apple systems later this year. It also can be set to run on hand-helds, computer kiosks and set-top TV boxes.

Shneiderman has launched a research program called Universal Usability (, which identifies the key problems to be solved: technological variety, user diversity and "bridging the gap between what users know and what they need to know." He is critical of software engineers who try to make computers "smarter" so that the machines will "know" what a user wants or needs at any given time--he thinks these automated schemes are guaranteed to be frustrating. Shneiderman points out that some companies are starting to get the message.

"Computers should be comprehensible, predictable and controllable," Shneiderman said. "Only then can the user be considered responsible for how they're used."


Gary Chapman is director of the 21st Century Project at the University of Texas at Austin.

Los Angeles Times Articles