Hard Cell

Design Principles: Of "Butt Phones" and "Theft Meters"

Of all the people working on the Smartphone project, Koerper, at 31, was one of the youngest. Just two years earlier, he had been rounding out his education with an MBA and a master's degree in electrical engineering from the Massachusetts Institute of Technology. But he carried himself with the earnest, relentless demeanor of a young Navy officer wanting to do well on his first command. Before coming to Kyocera, he had done everything from aerospace engineering in Los Angeles to microprocessor marketing in Japan. And as team members got to know him better, they discovered that the team leader with the wire-frame glasses and the close-cropped brown hair could be wickedly funny at times and a hell-raiser when the work was done.

Ironically, the Smartphone was built atop the rubble of a previous failure. In late 1998, something called the Qualcomm pdQ Smartphone was first test-marketed to major wireless carriers. The device did combine a mobile phone and a Palm platform in one device, but as one engineer wryly recalls, "It was a brick. It was big, clunky, and ugly." Customer feedback suggested that the design flaws were severe, perhaps even fatal.

So the word went out to people at Qualcomm's San Diego headquarters: Start thinking about how to produce a second-generation Smartphone that would be more appealing to customers. In December 1998, members of a small brainstorming team produced a 24-page "concept definition" for such a phone. Their memo was long on rhetoric about how vast the market for such a hybrid phone might be. But there were no sketches of what such a phone might look like -- just a single vague sentence that read, "Design should be optimized so the phone is comfortable to hold."

In other words, no one had any idea how to build it.

By a happy coincidence, Qualcomm had recently hired a cocky mechanical engineer from Hewlett-Packard, named Kyle Halkola. He was a dirt-biking enthusiast who was extremely proud of having helped HP design the casing for some of its printers. "I know plastics," Halkola would tell anyone who was curious. Eager to see just how good Halkola was, senior management assigned him to the Smartphone project.

For more than a month, Halkola, along with design specialist Rob Howe and members of an outside consulting firm, carved blocks of gray foam to produce prototypes of Smartphone designs. Their first dozen attempts misfired, particularly one dark-gray model with two bulging lobes in the back. They nicknamed it "The Butt Phone" and promised never to make something so ugly.

Top-level guidance came mostly in the form of oblique, Zen-like phrases. Koerper had read that Jeff Hawkins, the creator of the smash-hit PalmPilot, had tested his prototypes years ago by carrying them around in his pockets to see if they felt right. So everyone on the team conducted multiday pocket tests of each new version. Meanwhile, Paul Jacobs, then-president of Qualcomm's consumer-products division, urged developers to "make it look like a phone."

As Halkola and Howe kept plugging away, each of their designs needed realistic space estimates for all of the components that would go inside. Squeeze too much in one direction, and problems arose along another axis. When they tried to make a really thin model, it became unbearably wide. When they tried a narrower alternative, it got too thick. Then, around version 19, things began to click.

"You know you're getting close when every model you make starts to disappear," Halkola, 39, explains. "When you're way off, no one wants to touch your stuff. But when you're really on the right track, corporate vice presidents start to come by the lab and walk off with your latest version. We've got a name for it. We call it 'the theft meter.' And all of a sudden, our theft meter was being triggered."

Engineering Matters: Don't Be Surprised by Surprises

Meanwhile, serious work had begun on designing the guts of the phone. More than 600 tiny components would be nestled together on two circuit boards, each of which would be about the size of a business card. Randi de Lara took charge of the PDA; Mike Schnetzer oversaw the radio-frequency, or phone, section. The two engineers worked side-by-side for months -- and as they staked out space inside the Smartphone, they felt like two kids sharing the same small bedroom. When Schnetzer needed an extra 3 millimeters to accommodate a "tall" inductor mounted on his circuit board, de Lara had to help him out by agreeing to adjust parts on his board. Finally, after seven attempts to build the boards, both men had designs that were good enough to put into production.