Two Men, Two Visions of One Computer World, Indivisible

Two Men, Two Visions of One Computer World, Indivisible

New York Times, 1991-12-08

With computer technology advancing at a lightning pace, it's hard to imagine that anyone could work on the same project for more than 30 years — and still not finish it. But that is the case with two legendary figures of computerdom: Theodor Holm Nelson and Douglas C. Engelbart.

The two men are very different.

Ted Nelson is an outspoken author and visionary described as everything from "the Thomas Paine of the personal computer revolution" to a "madman extraordinaire."

Doug Engelbart is a soft-spoken engineer who invented the computer mouse, on-screen windows and several other techniques that are only now coming into widespread use. In a sense, Silicon Valley has spent the last 20 years implementing ideas Dr. Engelbart first demonstrated in the 1960's.

But Mr. Nelson and Dr. Engelbart have separately been pursuing very similar visions for three decades, and both know the frustration of falling short. "Trying to follow a crusading pursuit is not a very rewarding career path," said Dr. Engelbart.

The vision both have in common is called hypertext, a way of organizing vast amounts of information to be viewed on computers.

With paper documents, one usually reads from beginning to end in a linear fashion. With electronic information, it is easier to jump around because the computer can deliver any piece of information instantly. While perusing an article about Columbus, for instance, a reader might come upon a reference to Queen Isabella. With hypertext, the reader could press a button and jump to information about her, and from there to a passage about the Inquisition, and so on.

Hypertext is embodied in Apple Computer's Hypercard software and many other products. But both Mr. Nelson and Dr. Engelbart envision far more ambitious systems, linking networks of computers in which anybody could enter and retrieve information.

Dr. Engelbart, the engineer, calls his idea an "open hyperdocument system." Mr. Nelson, the wordsmith, calls his idea Xanadu, after Kubla Khan's legendary pleasure dome. "Once you're in Xanadu, the old computer world goes away," said Mr. Nelson.

Mr. Nelson has been pursuing hypertext since 1961, when he was a graduate student in sociology and wanted to organize his notes. He formed the Xanadu Operating Company to develop the necessary software in 1979. Some of the company's programmers have been working on it ever since, taking time off to get other jobs when money ran out but always returning to pursue the quest. Mr. Nelson kept saying Xanadu was close to being finished; it never was.

In 1988, Autodesk, a leading personal computer software company, bought 80 percent of Xanadu and put the company on a firm financial footing. Mr. Nelson became a "distinguished fellow" at Autodesk and lives in a houseboat near the company's office in this town near the Golden Gate Bridge.

The programmers, however, are kept 40 miles away in Palo Alto and now have little contact with Mr. Nelson. Sensitive to the credibility issue, they say a version of the software might be ready -- for testing -- in six months to a year. "We're doing our best to keep our heads down until we have something solid to show," said Chris Hibbert, who manages the software development.

Xanadu will be a database program to help people work together and keep track of documents. For example, an architect might be able to point to a spot on a blueprint and immediately jump to a memo explaining the reasons behind that part of the design. Someone reading a document that quotes from another document will be able to jump instantly to the source document. Users will be able to comment on something they read, and other readers will be able to jump from the document to those comments. People will also be able to put detectors into documents, to alert them automatically if someone subsequently comments on a particular passage.

A key to Xanadu is that each piece of information is stored only once but can be incorporated into many different documents, thereby saving disk space. On paper, if an author excerpts a passage from another person's work, he or she copies it, so that passage now exists both in the original and in the new work.

But in Xanadu, the new document only creates a link to the excerpt in the original document. One byproduct is that this allows the system to keep track of who is quoting whom, thus providing a basis for royalty payments on copyrighted material.

When, or if, Xanadu finally appears, Autodesk is expected to sell it to corporations. But Mr. Nelson wants to create a Xanadu for public use, to store and collect the knowledge of the world. That will require new funding and, the 54-year-old Mr. Nelson concedes, will take the rest of his life.

For the 66-year-old Dr. Engelbart, the idea is to help organizations make faster and better decisions. It was in working toward that goal in the 1960's, when he was at the Stanford Research Institute, that Dr. Engelbart came up with ideas like the mouse, windows and hypertext.

In 1977, Dr. Engelbart moved to Tymshare Inc., a computer service provider that sold his system under the name Augment (as in augment human intelligence). But Augment was offered on big computers, and business faded when personal computers proliferated.

In 1989, Dr. Engelbart set up the Bootstrap Project to try to interest companies in using his techniques to "bootstrap" themselves toward better performance. Initial funding came from Apple Computer, Sun Microsystems and Mitch Kapor, the founder of the Lotus Development Corporation.

But that funding ran out and, partly because times are tough, partly because Dr. Engelbart's ideas are somewhat difficult to understand, has not been replaced. He now runs Bootstrap with his daughter Christina Engelbart, operating out of a donated office in Fremont, Calif., and trying to keep the project alive with seminars and videotapes.

It might seem that Dr. Engelbart and Mr. Nelson could join forces. Indeed, Dr. Engelbart thought of using Xanadu as the underlying software upon which he could build a new version of Augment. But he could not reach an agreement with Autodesk.

Both men now face advancing age and competition from numerous other companies already producing software that helps workers collaborate and keep track of documents. Lotus's Notes program is one example.

Both Mr. Nelson and Dr. Engelbart think these solutions fall short of their ultimate visions. But perhaps that is how the technology is destined to develop, piece by piece rather than in one all-encompassing leap. In that sense, these developments by others are not so much a threat to the two pioneers as a sign that their visions will be realized, though perhaps not by themselves.

"Doug is like Moses," said Paul Saffo, a researcher at the Institute for the Future in Menlo Park, Calif. "He's leading us toward the information Promised Land but he will probably never enter it himself." Mr. Saffo also doubts that Mr. Nelson's Xanadu will ever be finished.

"Neither will probably ever reach their goals," he said. "But even if they fall short, they fall forward, and all of us benefit."

Douglas C. Engelbart and William K. English: A research center for augmenting human intellect, AFIPS 1968-12 (PDF)