Don Norman's aim is to make companies produce more user-friendly products. Tony Durham met him

Don Norman cannot ignore a badly designed door, tap or light switch. He is likely to photograph it, figure out why it is unfit for human use, and put it in his next book. Everyday life must be pretty annoying for this man, one imagines. On the other hand, events that might have been major frustrations in his career do not seem to impinge on the contentment that beams from his face.

In 1993, after years at the University of California, San Diego, the electrical engineer turned psychologist decides academics are out of touch with the real world. He does not agonise over wasted years. He quotes the management writer Tom Peters: "If you are very happy in your job it is time to move on." He calls some friends in industry, takes the university's offer of early retirement and lands a job as an Apple Fellow, one of the computer company's senior researchers who can pick their own projects.

He wanted the real world. He gets it. Apple is losing money, and a new management decides research is a dispensable luxury. At times like this, it helps to have a reputation. In the admittedly confined world of human-computer interaction, Norman has one of the best. His 1986 book User-centred system design, co-edited with Steve Draper, "defined the field of human computer interaction", according to Harold Thimbleby of Middlesex University, himself a noted critic of ill-designed machines. Norman's second book The Psychology of Everyday Things became a best-seller.

É«ºÐÖ±²¥

Without much effort, Norman finds a welcome at another big computer company, Hewlett Packard. Once again the future looks bright. Everyone he meets, including the company's research vice-president Joel Birnbaum, shares his interest in information appliances - simple, specialised devices such as cellphones and palm-sized electronic organisers. Unfortunately, it turns out he has met the half-dozen information appliance enthusiasts in a company with more than 100 divisions. The ink-jet printer division is locked in battle with the laser printer division. Neither is particularly receptive to wild new ideas from the labs.

Once again Norman is stuck in the role of design critic. He would have liked to lead Hewlett Packard into the age of the information appliance, building on its experience with calculators, scientific instruments and gadgets that talk to each other by infra-red beam. Instead he writes another book, The Invisible Computer, the book he is in London to launch.

É«ºÐÖ±²¥

When it comes to exposing the design flaws in industrial products, Norman is peerless. For former Apple technology vice-president Larry Tesler, "Don is a superb and acerbic critic of machines and software that are confusing to use."

As a critic, Norman likes to be constructive. He showed the draft of The Psychology of Everyday Things (POET) to a group of industrial designers, who felt affronted by it. He rewrote the book. "It is actually sympathetic to the design profession. I talk about the challenges they face. They don't have full control over the product" What Norman gave back to the design community was a psychological explanation of why some designs are usable and some are not. "It was one thing to rail about how badly things were designed. I also wanted to have a positive contribution and try to point out what the principles really were. And I sat down asking myself the question, how is it that I get through everyday life without having had most of what I do explained to me? How do I know where the door is and how to open it? How do I understand everyday life was the question, and that is what led me to affordances. As I looked around it became clear that most of the information was in the world. It did not have to be in the head."

Here he drew on the idea of the psychologist James J. Gibson, that every object has a set of affordances, or things that can readily be done with it.

Gibson was concerned with the practical affordances of natural objects. Norman had in mind something more artful, a set of perceptual cues that a designer could use to influence our behaviour. Doors are his favourite example. Some have handles which ask to be pulled, others have plates and bars which demand to be pushed. Only the designer who has failed to provide these cues has to resort to "push" and "pull" signs.

POET sold slowly at first. In paperback it was renamed The Design of Everyday Things so that bookstores would shelve it with books on design, not psychology. Sales perked up. And with them, surely, the author's reputation? "You don't understand about the academic community. It was a popular book. Nothing is worse than one of their members writing a popular book. Look at Dawkins. You think he is respected for his popular books?" But Norman had discovered his gift for popular writing, and he added Turn Signals Are The Facial Expressions of Automobiles (1992) and Things That Make Us Smart (1993) to his credentials as the Dawkins of design.

Now, in The Invisible Computer, he has rounded on that 1990s object of desire, the personal computer. "Basically the PC is hopeless," he told an audience at Imperial College, London. "There is no way to fix it." It is too complicated, "evil, even". The good news: the work we used to do on PCs will be accomplished on a variety of information appliances, mostly small and without trailing wires. "You will find you are using your PC less and less and at some point you will just stop using it."

É«ºÐÖ±²¥

The term "information appliance" was coined in 1978 by Jef Raskin, originator of the Apple Macintosh project. But it has taken manufacturers years to create something as simple to understand and use as the 3Com Palm III, a curvaceous steel clamshell that rests in your palm, understands your handwritten scrawl and quietly organises your life. Norman pulls one out to look up an address.

Future appliances may be dedicated to domestic finances or school homework. But the gadget most likely to give academics pause is a book-sized box with a screen on the front and enough chips inside to store a small electronic library. Two book-readers, the Softbook and the Rocket eBook, are now on the US market.

É«ºÐÖ±²¥

And Norman has just visited a bookstore in Massachusetts that prints technical books on demand. "They can print a book in 30 minutes or less, and they bind it."

The frustrations of Norman's industrial years have left their mark on his own book. Something has clearly gone wrong when computer company helplines are blocked with calls from desperate users. If anyone has the right to say "don't blame me, I warned you", it is Don Norman. But he resists the urge to blame Microsoft's Bill Gates or anyone else, arguing instead that all technologies - radio, the telephone, the phonograph - are fiendishly complicated in their early years and the PC is no exception.

The way Norman now presents it, usability must find a compromise with engineering, marketing and manufacturing. Has usability's champion sold out? "It doesn't matter if you make the world's most usable products, if nobody buys them. It is not selling out. I want to make sure that people buy them, which means understanding what causes people to buy."

In industry he met clever people whose concerns were different from his own. If he could not convert them, at least he could understand them. "I believe I understand the business model, the marketing model, the selling model, the design model, the usability model. Can't we put them together?" And this is what he is now trying to do.

He has left Hewlett Packard and joined forces with Jakob Nielsen, another top usability guru who was encountering similar frustrations at Sun Microsystems. If they were rock stars this would be a supergroup. The Nielsen Norman Group is the Emerson, Lake and Palmer of human-computer interaction, and its supposedly quiet launch two months ago was swiftly reported in Business Week, Forbes and Silicon Valley's local paper The San Jose Mercury News.

The idea was that Nielsen and Norman would take the design message to senior management in big companies, where it is rarely heard. Instead, things are unfolding in typical Silicon Valley fashion: "We have been approached by several startups, and they are paying us in stock." Big business is slower to come to the bait. But there is plenty of time. Academia may have been unreal, and industry frustrating, but Don Norman's third career has only just begun.

É«ºÐÖ±²¥