Apple should quit the mainstream PC market and concentrate instead on multimedia production and entertainment, the company's former design head has claimed.

Noting that there were "some things I can't say", Apple's design guru in the early 90s, Don Norman said that the computer company's corporate culture and over-emphasis on creativity was the reason behind the Mac failure to win the computer market, leaving the way open instead for Microsoft and IBM.

"The business story is rather complex," he said. "Apple became arrogant and didn't treat its customers well. It rewarded creativity, which meant it brought out some brilliant products, badly. Things about the Apple designs needed fixing, but nobody wanted to fix anything, because that wasn't creative.

"And one point where Apple really shot itself in the foot was to come up with the message that the Macintosh was 'The computer for the rest of us' - not for you dull business people. So Microsoft and the IBM-style PC, not surprisingly, sold very successfully into the business market."

Norman has since worked for Hewlett-Packard and is now advising Microsoft on the design of its upcoming Windows OS, Longhorn, which should give Mac advocates a useful conspiracy peg to hang his words on. However, as a man that has seen - and helped direct - the computer market since its early days, his words carry some weight.

And he is continuing to look into the computer's future. The world wants compatibility now, he says. It wants to communicate, and this means one brand dominating. "This is not the 'computer age' any more; this is the age of very smart chips hooked into a huge worldwide network. Infrastructure is about sharing." This means two or more incompatible ways of doing things is counterproductive.

As he forsees a radical change in how we actually interact with computers. The classic graphical user interface that we all currently use - files in folders - was well suited to an early Macintosh with 128KB of RAM, he said, but it doesn't scale.

With few tasks in those days, the GUI was excellent. "You didn't have to remember anything, because you could see everything. Now making everything visible doesn't work. The space gets too crowded." As a logical consequence of this, the all-purpose computer should become obsolete, he says.

He now sees a machine based on a searchable database, doing away with a static file system view. The document-folder-cabinet hierarchy may be a fair simulation of the way an office works, he says, "but I just want to get my work done". Wanting to send an e-mail with an attached document and having to look in different folders and maybe start up a word-processing program is an obstruction to the natural way of working. It clearly makes sense to store everything once. Conventional file systems tend to produce duplicate files stored in different places.

Norman still thinks highly of many features of Apple design, however. He cites the functional aspects, to the "pleasant experience" of unwrapping the machine from the box, and "the illuminated logo that tells everyone you're an Apple user and gives you the feel of belonging to a community".

Good design succeeds on three levels, he says: the reflective - the impression of the product on the intellect; the behavioral - the way you use it; and the visceral - its appeal to the emotions. Sadly, the modern world is less intent on emotion and more on efficiency and profit.