Unravelling technology's riddles

By ADAM GIFFORD

"The process of invention is often playful, but luck favours the well prepared."

That is how Dr Ross Allen, from Hewlett-Packard Laboratories in Palo Alto, California, explains how he has accumulated a string of patents for technologies such as the HP CapShare portable document scanner, low cost Machines for making books and booklets, methods of identifying what kind of paper is in the printer and a device that measures the amount of toner left in a LaserJet cartridge.

Dr Allen, the manager of HP's colour imaging and printing technologies department, is in New Zealand giving technology briefings to HP staff and resellers before taking a holiday in the South Island.

In 1981, the mechanical engineer with expertise in electrical engineering and computer science, left a promising academic career to join the team of HP researchers developing the recently discovered thermal ink jet technology into a commercial product. His inventions have come from chance connections as well as formal team processes.

The laser toner idea was the former. "I was sitting in a review at HP Labs with our LaserJet folks from Boise, Idaho," he remembers.

"A senior manager said 'We'd like to know exactly how much toner is left in the cartridge.'

"I looked up from my desk and said 'I know how to do that'."

A year before that, Dr Allen had seen a demonstration at the Massachusetts Institute of Technology of capacitive field sensing.

When a person moved his or her hand between two metal plates, changes in electric charge were measured so that the position of the hand could be displayed in three dimensions on a computer screen.

While the demonstration was interesting, it had no immediate use, so Dr Allen filed it away until the request from Boise.

"I said 'Toner is plastic. Plastic is a dielectric [or insulating substance]. If you have a dielectric between two metal plates, if you know the capacitance you know exactly the amount of dielectric in that geometric orientation.'

"We went into the lab, prototyped it the next day, and six months later had a working solution that could probably be delivered for $1."

Compared with that flash of inspiration, CapShare, the invention Dr Allen is proudest of, came out of team brainstorming.

"It came from a group of people who would meet every two weeks in a room with a white board with the idea 'We're going to invent a new product - we don't know what it is, but it will be something aligned with HP business.'

"Someone said 'Wouldn't it be nice if we made a polka dot paintbrush, something you just wiped across a surface and it worked like an ink jet or laser printer, but you could move it in a freehand way?'

"I said 'That's a very hard problem, because as this printing array moves in a plane and rotates, it's hard to know where these dots are going to be. But if you make a scanner that way, you already have the information completely correlated, and all you have to do is discover it by knowing where this thing was.'

"Then we said, 'How do you deal with white space where there's no image?'

"And someone said, 'If you look at the structure of paper, on the level of say 40 microns where paper looks like a bale of straw, and correlate those images, we will know exactly where we are when we move through space.'

"From that, a year later, we had a working prototype."

CapShare has a pair of custom chips, called Navigators, which do over 3 billion operations a second, processing 10,000 images a second and correlating each image with the next.

Despite being a technological tour de force, CapShare failed in the marketplace, as few people were willing to pay $US700 for a portable scanner.

"We could have done a better job of understanding who the customer was and what they were prepared to pay."

When CapShare was "retired," Dr Allen went to HP's chief executive.

"I said 'This is a unique technology, it's too good to lose, we have to find the customer and get it right'."

The result is CapShare will be relaunched "soon" at about $US250 and with greater usability.

Dr Allen says the goal of HP Labs is to invent the future for the company.

"We look anywhere from two to seven years into the future, figure out where the business is going and where the unmet needs are, look at things which are expensive and inconvenient to do today, and find new technologies.

"We typically look at what people are doing now with $15,000 to $100,000 solutions and say 'Is there another way of doing that?"'

Much of his work now involves fast DSP (digital signal processing) chips, which do away with the need for custom chips on devices and therefore radically shorten invention cycles.

Dr Allen says there is still a place for small inventors who do not have the resources of a $40 billion company behind them.

"You have to match technology with need.

"For the small inventor, history is full of people who are successful with a simple, good, compelling idea. You don't have to sell it to a big company, but you have to find a way to get it into the hands of other people, so need a network of some kind."

HP's Ross Allen