Reprinted from Popular Computing, issue 5/84, pp. 70-78.

Illustration by Richard Cowdrey

Not more than a hundred yards away from the complex of buildings in Cupertino, California, that house Apple Computer Inc., the company that popularized a particularly useful pointing device called the mouse, is the headquarters of another high-technology firm named Tymshare. Working in that building, heading a little-publicized project, is a gentle, soft-spoken gray-haired man who looks to be in his late fifties. His name is Doug Engelbart, and he is the inventor of the mouse.

I had come to visit him on other business, business having little to do with rodents. But it had not been long since the introduction of Apple’s Lisa computer, and the irony of Engelbart’s proximity to Apple’s home territory had not escaped me.

One of the much-touted features of the Lisa is the mouse. Apple does not claim to have invented this enchantingly effective pointing device, but its corporate sales pitch makes it clear that Apple considers Lisa’s mouse a giant step in the annals of productivity. By shoving around this piece of hardware the size of a pack of Camels, you can move a cursor on the display screen; by pushing its button (the Apple mouse, unlike previous mice, has only one button – more on this heresy later) you can select items on a menu to execute commands. Yes, there had been mouse technology at Xerox’s Palo Alto Research Center (PARC) and, yes, some high-priced computers like the Xerox Star and the symbol-crunching LISP machines use mice. But in a real sense, Lisa marked the debut of the mouse for the general public.

Mouse fever

Sure enough, in 1983, the mouse became a craze. Besides Lisa, we saw mice offered by Microsoft and Visicorp for the IBM Personal Computer. By the end of the year mice were matched to the Apple IIe, and this year’s Macintosh comes equipped with a mouse. Sometime last year it became apparent that the mouse was going to be standard equipment on many computers. What did not become apparent was the fact that the mouse is just a tiny piece of a large, ambitious plan conceived by a truly extraordinary man. I thought I should tell you about this man and his vision.

Douglas C. Engelbart was in his shirtsleeves. He greeted me warmly and walked me to his office, a cubicle in one corner of a large room filled with file cabinets and cubicles. We sat down to talk about the quest he set out on over thirty years ago. He had been working for the government agency that would eventually become NASA, doing some advanced research with big wind tunnels – a great job for a 25-year-old scientist.

He had just become engaged. Yet he felt something lacking. “I was going to get married, and I had a good job at an interesting place, and suddenly it all seemed much too flat to accept” he recalled. “So for the next weeks and months I tried to feel what could be a goal... a career commitment that would really make a difference.” As wild as it seems, Engelbart wanted to do something that would have a direct impact on shaping the future of humanity. Sitting there in his open cubicle at Tymshare – one of those quirkless and thickly carpeted office complexes where conversation floats through space in a dull drone – he referred to the work he decided on as a “crusade.” Instead of tackling a specific problem, he decided to go about changing the way we dealt with problems, the better to improve the world. He saw that many problems were so complex that it was often beyond human capabilities to solve them. Somehow Doug Engelbart decided to improve – augment is the word he came to use – man’s capabilities to cope with those problems. The logical way to do that was by applying new technology and developing ways to interact with that technology.

Lonely work

Engelbart admitted to me that when one enters the crusade business, “you don’t have the tried-and-true sort of formulas for getting involved. You can’t find a marketing and production manager and all that.” At first the work is lonely. He knew he would have to learn much about computers to pursue his vision, so without hesitation he quit his job and took his new bride to the University of California at Berkeley, where he earned a PhD in the young field of computer science, and then got hired by the Stanford Research Institute (SRI), hoping to work on his crusade from there. It took a while to get anyone interested in freeing him to pursue his vision, but he eventually won a small grant, which allowed him a few years’ time to come up with a framework to change the way people work and think.

By early 1963, he had the framework. This was 1963, mind you. Computer science as we know it did not exist. The idea of plain people using computers – using them to perform the kinds of nonnumerical “thinking” tasks they did in offices – was somewhere this side of science fiction. And here was this guy Engelbart publishing a paper called “A Conceptual Framework for the Augmentation of Man’s Intellect.” The prose was unemotional and proceeded evenly. But the ideas in it were virtually unheard of.

Basically, Doug Engelbart was saying that our problem-solving capabilities – the way we’d been using tools, interacting with tools, and especially handling information – were unnecessarily limited. “Augmenting man’s intellect” would result in better comprehension of problems, quicker solutions of problems, and solutions of problems that were previously unsolvable.

This was nice to say – we would all like to have the powers of Superman – but Engelbart talked as if it could be done. And he hinted how – by a system using digital computers, where everyone would have access to a computer workstation, be in contact with each other, and even compose memos and documents by a strange process in which “trial drafts can rapidly be composed from rearranged excerpts from old drafts... you can integrate your new ideas more easily and thus harness your creativity more continuously.”

This sounds a lot like what we now call word processing, certainly one of the more revelatory changes in our work lives. But this was only one of the innovations Engelbart was hinting at, to create a system to further this very large revolution which, he later proclaimed, would have a bigger impact than “the combined effects of the printing press and the industrial revolution.”

“That paper was the first time or place I came out with those thoughts,” Engelbart told me more than twenty years later. “There was a lot of risk there.”

I asked him what the response had been to that ground-breaking manifesto.

He looked at me unblinkingly. “There was one little review someplace that said, ‘Here’s a description of a little documentation system.’ ”

That was all. Near silence. It was to be a recurrent theme. Although Engelbart obtained some funding from the Air Force and the Defense Department to start an Augmented Human Intellect Research Center at SRI, the science establishment never really took him or his work too seriously. Even as his research team started to get results, began to develop the technology and methodology that in some way resemble our currently most advanced vision of The Automated Office of the Future, he received little notice from the science press and “neutral to negative” response from the academic computing community. He explains: “The perception is very different if you’re trying to make better tools in a human framework instead of trying to make machines smart.”

Some of those tools were conceptual – enabling people to jump from idea to idea without obstacles. Only recently have most of us become familiar with the concept of “windowing,” which lets us see different things on the display screen at one time and move data from one file to another. Engelbart had it years ago.

Other tools were more simple physical aids to using computers. For instance, in the mid-sixties, his group, working out of barracks-like World War II-vintage huts on the grounds of SRI in Menlo Park, had terminals unlike any others in the world: display screens standing alone, with separate keyboards. And external input devices for both hands. For the left hand, there was a five-key “handset” that looked like a variation of the foot panel of a piano. For the right hand, there was a mouse.

It is odd that the mouse should be the most celebrated of Engelbart’s achievements. Compared to the scope of his goals, it is a minor advance, something that allows easy “pointing” to something on the screen. The group tested various methods, including joysticks, track balls, light pens, and even a weird knee-controlled device. None worked as well as the mouse, later described in a paper of Engelbart’s as a device that is of “palm-filling size, has a flexible cord, and is operated by moving it over a suitable hard surface that has no other function than to generate the proper mixture of rolling and sliding motions for each of the two orthogonally oriented disk wheels that comprise two of the three support groups... That the mouse beat out its competitors... seemed to be based on small factors: it stays put when your hand leaves it to do something else (type or move a paper) and reaccessing proves quick and free from fumbling. Also, it allows you to shift your posture easily, which is important during long work sessions... And it doesn’t require a special and hard-to-move work surface... A practiced, intently involved worker can be observed using his mouse effectively when its movement area is littered with an amazing assortment of paper, pens, and coffee cups, somehow running right over some of it and working around the rest.”

Its moniker came from the humped, palm-fitting curve of the device, along with the way its cord looked like a tail. Also, if looked at in a certain way, the three identical buttons seemed to represent a nose and two tiny mouse ears. “We just nicknamed it that,” says Engelbart. “We never really thought about it in the way of sticking with it.”

I later talked to one of Engelbart’s assistants at the time, a computer scientist named Bill English, about the development of this pointing device. English helped design the prototype. Among the factors they looked for were comfort and control. “You want it fairly low, without your wrist on the table. Ideally, you’d like to use the same muscles you use to control a pencil.” A woodworker custom-built a frame, and a trial-and-error process decided the size of the two wheel-like discs on the bottom which were attached to a potentiometer in the mouse. The rolling action of those wheels would thereby be registered in the machine and translated to cursor motion. The first model had a single button, but that was quickly changed to three. “Clearly, three was the right number,” English told me.

That statement launches us into a digression from Doug Engelbart’s story and into the How Many Buttons Brouhaha, which examines why such a computer luminary as Microsoft’s chairman, Bill Gates, recently said, with a certain wistfulness, “The number of buttons on a mouse is one of the most controversial issues in the industry. People get religious.” Bear with this digression; we will return to Mr. Engelbart’s vision forthwith, with an explanation as to why this fork was taken.

Bill English went to Xerox PARC in 1971, and among his goals was the attempt to bring some of Engelbart’s innovations to the marketplace, including the mouse. But it was not clear at all that three buttons was the right number. Xerox did a new set of tests, which generally confirmed the virtues of the mouse, uncovered the fact that “small is better” when it comes to mice, and found that less than 10 percent of users had trouble grasping the concept of mouse use. (For those 10 percent, Xerox devised a program called Fly, in which a person used the mouse to control a fly swatter on the screen: he or she would use the swatter to squish “flies” on the display and “after 10 minutes, they would be perfectly at home with the mouse,” English says.) Xerox found that instead of three buttons (which some thought were too many) two might do just as well. “Two is a good compromise,” English says. “With one, you have to go through a lot of contortions to select things.”

So the mouse on the Xerox Star system, and later its Altos line of computers, had two buttons.

But then came Apple’s Lisa, and a Xerox PARC alumnus named Larry Tesler was one of its designers. Since, as he explains, “our interest was getting systems to laypeople, not computer experts,” he thought that a multibutton mouse was overly complicated. “People made a lot of mistakes. With two buttons, they’d constantly be turning their heads from the screen and looking at the mouse. When we did experiments with identical everything, except the number of buttons on the mouse, the people who used a one-button mouse said it was easier to pick up. I didn’t have any religious aversion to one button,” so Tesler argued for it.

Bill English thinks Apple’s mouse “was a mistake. Why do you want to limit yourself like that?”

Tesler insists that one button is not limiting. “We’ve never regretted it and never received a letter from a Lisa user asking for another button on the mouse. There are some two-button advocates here, but then there’s the question of what you use the second button for. Some people use it for, canceling, some for scrolling.” Tesler reiterated that the one-button mouse is more intuitive, easier to learn, and with the double-clicking ability can be as versatile as a two-button mouse.

English is not convinced, nor, apparently, is Bill Gates (whose Microsoft mouse has two buttons), nor are the Visi On people, who also go with two. Apple’s Macintosh team re-tested the whole concept of the mouse before they went with it in their machine (including one golf-ball-shaped mouse nicknamed The Arnold Palmer Model) and stuck with Lisa’s one-button concept.

Sitting there in Doug Engelbart’s office, I heard a lot about the relative virtues of easy-to-learn systems versus more complicated, richer ones. The elaborate augmentation system that Engelbart has crafted, of which a mouse is only a small component, is not easily mastered. Not only does the mouse have three buttons for the right hand to push, but there is also a left-hand “keyset” with five keys, which can be pressed one at a time or in “chords.” I watched Engelbart demonstrate the tools, moving text on the screen, opening windows, and calling remote files. At one point, with his left hand on the keyset and his right on the mouse, he managed to write an entire line of text without touching the keyboard. He did it without moving his eye from the screen. It gave me the illusion that the computer itself had written the line.

This requires considerable deftness. Unlike a one-button mouse, you can’t master this stuff in 10 minutes. But Engelbart offers no apology for this. “I say, ‘Hey look – do you ride a bicycle or a tricycle?’ Obviously, a tricycle is easier to master. But how fast can you go? It might take some time to learn how to ride a two-wheeler, but once you’ve done it, you can zip through streets like a sleek master of your environment.”

Thus the mouse controversy is symbolic of our resistance to any new method of working, especially one as complex as Engelbart’s. Will systems be limited because we refuse to take the time to learn their complexities?

As I watched Engelbart use his augmentation system, I kept that in mind. His workstation, with the exception of the exotic left-hand keyset, was familiar to me as a model for the office of the future. Of course, having worked in this configuration for over twenty years, Engelbart handled the tools like a master.

The real augmentation

As Engelbart explained to me, the physical tools are only the access points to the augmentation system. The heart of it is a database concept. Your computer is hooked to local networks and distant computers with voluminous data files. All text is marked and flagged, and the system is set up to get information on a subject whether the information resides in a letter, a memo, a technical document, or a book. The vision is to have someone sitting at a workstation using Engelbart’s tools to race through the collected wisdom of humanity, wisdom that will be used to solve problems faster. Or to solve problems that wouldn’t have been solved before.

“I’m really going after high-performance people,” says Engelbart, “people who will take a system to the hilt. If some program could achieve a singularly unique level of effectiveness, even for people who are highly trained, it would awaken perception that it’s there to go after. What generally changes people’s perception is experience.”

The augmentation system is being offered, on a limited basis, by Tymshare, which bought Engelbart’s research and, in effect, Engelbart, from SRI in 1977. “There was a slightly less than universal perception of our value at SRI,” Engelbart says with a terseness that might once have been frustration or anger. Engelbart and his people were given a chance to raise money to buy the group from SRI; they could not interest the venture capitalists. So SRI put the project up for auction, and Tymshare got it for a bid representing a fraction of the accumulated research costs. “A cheap way to take a chance,” says Engelbart. “It wasn’t due to an understanding of the augmentation side of it.”

The “augmentation side of it” is what sent Engelbart on his quest several decades ago. It pains him that many of those who recognize him as an innovator do so only because of the mouse. “Everyone translates what I say into the framework as they now perceive it. ‘Lisa and the Star are here now,’ they say. ‘Now that everyone is doing what you want them to do, why don’t you relax about it?’ ”

He sighs, and this time the weary frustration is obvious. “I guess what I really want is that conscious pursuit of the whole system of methods and skills and languages.” So the world will be changed, and man’s power extended. Augmented.

He looks down at the implacable mouse to the right of his keyboard. The mouse that has won the pointing device sweepstakes, the mouse that has taken the thunder from the marvelous whole system of augmentation he has spent a lifetime to develop and promote.

“Now you might understand,” Engelbart says, “why I wrinkle my nose so much when I hear that a mouse is ‘easy to learn and natural to use.’ ”

by Steven Levy

Steven Levy is a contributing editor of Popular Computing.