Imagine a handheld gadget that's a phone when you put it to your ear, a camera when you hold it to your eye, and a game device when you grasp it on either side. A single gadget that changes function on the fly based on how you hold it is the idea behind the Bar of Soap, a prototype handheld built by Brandon Taylor. Taylor is a graduate student in the Massachusetts Institute of Technology (MIT) Media Lab, under the direction of Professor V. Michael Bove, Jr., the director of the MIT Consumer Electronics Laboratory.
Graspables
MAJOR INNOVATION A device's function—camera, phone, PDA, game—is determined by the way it's held.
WHY IT MATTERS With enough information, a device can learn to conform to the user's needs instead of the user adapting to the device.
ESTIMATED ARRIVAL With corporate interest, within a year or two.
The Bar of Soap is an example of what Bove calls graspables: "things that could be handheld—a steering wheel, a baseball, whatever—that knows how you're grasping it." He foresees a future where objects all around us are aware of how they're being used. For example, an object you're holding could recognize when it's aggravating a medical condition, like carpal tunnel syndrome, and adjust itself to compensate. Or when you pick up a mug, your Second Life avatar also picks up a mug.
Taylor calls the initial Bar of Soap a "mode-sensing, multi-function device." Inside the rudimentary prototype, an accelerometer measures how the device moves. Outside it has a touch screen and is covered with 72 capacitive sensors that detect the hands holding it. "We trained users to hold it in different ways," says Taylor, "and saw what orientation it was in for different modes of use."
Naturally, not everyone holds a product the same way, not even a phone. The Bar of Soap records how it is used to determine how a population holds the unit in phone (or camera, or gaming) mode. It can distinguish when it's used as a phone rather than, say, a remote control (another function built into the Bar of Soap prototype, along with a PDA and some games). "It has fairly sophisticated pattern recognition for a cheap device," says Bove.
Obviously, you would want such a device to adapt to the particular way you hold it. That'll be easier when the hardware is seeded with a large data set of variations of how it can be used. Taylor points out, however, that users who didn't get the function they wanted with the prototype adapted the ways they held it to bring up different functions. "We wanted to see how people hold things and how it carried across different users. Most people don't want to do much to change from one function to another. With this device, "you don't have to go to a camera menu. It just knows. Or it knows you're making a phone call. It doesn't bother you with irrelevant other stuff," he says.
Taylor will have graduated by the time this sees print, having completed his thesis on—what else?—graspables and usage pattern recognition. He thinks the Bar of Soap is just the first step. "The next jump will be the graspables idea in different geometries, not just a multifunction handheld."
The MIT Media Lab has shown lots of industry sponsors this first graspable and discussed its possibilities, but no products are on the horizon yet. This is almost inexcusable, because the Bar of Soap prototype doesn't involve any new or special technology—just the smarts to utilize it. "The algorithms and components are available now," says Bove. "You don't need five pounds of Unobtainium to build this. It uses principles feasible for products right now. This is potentially quite close to market."
