Don Norman: Designing For People

Nielsen Norman Group

A Time For Standards

Originally published in InteriorMotives

There comes a time in every product’s life when it is time for standards. Ah, standards, the bane of our existence – as well as our salvation. Standards that come too early lock in old-fashioned technology preventing further advance. Standards that come too late, well, are too late to make a difference. Market leaders traditionally oppose standards: market followers embrace them.

But now is the time to standardize the ever-increasing use of automation within the automobile cockpit. This is a safety issue. It is not about sales or market share, it is not about stopping innovation. Indeed, I believe that proper standards, adopted now, might encourage innovation and, more importantly, customer acceptance.

Consider the following scenarios, two out of many possible issues to be faced. In both cases: What actions should the car take? How should it signal you?

One. You are driving along and your car senses a potential crash ahead.

Two: Your car senses that it is drifting out of its lane.

Systems have already been deployed for both these cases, using different philosophies to signal the driver. For impending collisions, in some cases, the car itself prepares for a crash, straightening the seats, tightening the seat buckles, adjusting the shocks. In some cases the car buzzes, vibrates, or beeps at the driver. In some cases the car brakes automatically, sometimes after viewing the driver with a video camera. In one case, the car pushes the driver’s foot with the accelerator pedal. For lane drifting, some cars vibrate the tires on the offending side, some vibrate the seat. Some beep, some vibrate the steering wheel. And one company applies 80% of the corrective torque necessary to get back into lane, but only if the driver supplies the remaining torque. If the driver does not respond, the system disconnects.

There are legitimate reasons for each of these philosophies. Problem is, they are all different. But today’s drivers, especially the affluent drivers who are the most likely to have purchased the high-end vehicles on which these systems are deployed, drive multiple cars. Their family may own multiple cars, their company may provide them with a car, and while traveling, they experience a wide variety of rental vehicles. What happens when drivers who are accustomed to one philosophy of warning encounter another philosophy? They might very well respond inappropriately, increasing the severity of an impending accident.

One automobile executive has already told me of one such incident. His company provides him with a different vehicle to drive every month. After several months of experience with different cars that had backup warning systems that signaled the distance remaining to objects behind the car, he was given a car which lacked the warning device. Oops. One family member had come to rely upon the warning, and in its absence, backed up through the rear of the garage.

Night vision systems are being deployed, but with different kinds of underlying technologies, different locations for the displays, different kinds of warnings and, because of the differences in the underlying technologies, different sensitivities to potential road hazards. Automatic parking systems are now available. Adaptable cruise control systems now operate at rush-hour, city traffic speeds (that is, all the way to stopped traffic), but each requires different methods of resuming control after a stop. The list of potential issues grows larger with every new model release.

Standards committees face daunting tasks. Nonetheless, the time has come for this one. If we do not standardize the nature of the warning systems, we are apt to create as much harm as good with these systems. We need functional standards that specify the forms by which signals might be given, the nature of the automatic operations that will be performed, and to standardize the controls and displays, so that even drivers new to a vehicle can understand the warning, know what actions to expect, and know how to read the displays and modify their settings. Today, we have a hodge-podge of methods. Hodge-podges and safety do not go hand in hand.


Donald Norman is co-founder of the Nielsen Norman group, a psychologist/cognitive scientist/design theorist who teaches at Northwestern University and, in his spare time, writes books, including “Emotional Design: why we love (or hate) everyday things.” He lives in northern California at www.jnd.org. Write him at don@jnd.org.

Essays

Books