Three Challenges for Design

Column written for Interactions, volume 14, issue 1. © CACM, 2007. This is the author's version of the work. It is posted here by permission of ACM for your personal use. It may be redistributed for non-commercial use only, provided this paragraph is included.

The invisible, ubiquitous computer has arrived, ensnaring almost any conceivable activity within its grasp. This raises wonderful opportunities and challenges to the field of human-computer interaction, for if the computer is everywhere, then everything is within our domain of study.

It is time to consider where the next application areas might be. As I look to the future, I see numerous domains of concern, but with three large, overriding issues:

The ever-increasing complexity of everyday things
The ever-increasing burden of security, authentication, and identification
The ever-increasing use of automation

Obviously, this list of three hardly covers everything. After all, if the field covers almost everything, the list itself should be of almost everything, from voting machines to educational systems, finance to entertainment, health to child care. But these three are broad enough to keep us busy for many a year.

The ever-increasing complexity of everyday things

As our demands for services, functions, and features grow, so too does the complexity of our devices increase beyond reasonable comprehension. Cameras and cellphones, audio and television equipment, kitchen appliances and automobile dashboards. Some home toilets now require electricity, display panels, and instruction manuals. This complexity often leads to frustration, and sometimes poses serious safety hazards. In particular, automobile dashboards require multiple button pushes and glances at visual menus, dangerous while driving. Mobile phone usage is a known danger, with at least one study showing it is more dangerous to drive while talking on a mobile phone -- even hands-free phones -- than to drive while legally drunk (Strayer, Drews, & Crouch, 2006).

Although the science of complexity reduction is well understood, the practice is not. Automobile cockpits have special requirements because their controls must work well with unskilled people, under stress, in limited time, where the device is usually not the major focus of attention. When a driver wishes to change the interior temperature, or change the radio station or music selection, this activity should not distract the driver from the primary task of driving. This means that the instrument panel, already crowded with displays, navigational devices, controls for multiple passenger temperature setting, driving parameters, and entertainment must be able to be used by someone relatively unskilled at both the task at hand and even the primary task of driving. Moreover, the eyes should not leave the road for more than two seconds. Why two seconds? Both common sense and the results of a well-documented field trial show that accident rates increase sharply as time the gaze is off the road exceeds two seconds (Klauer, Dingus, Neale, Sudweeks, & Ramsey, 2006).

What's true in the automobile also applies in the home, although without the same time stress and safety concerns. Appliances are increasing in complexity. Washing machines and driers, dishwashers and microwave ovens, coffee makers and refrigerators are all now available with complex menus, multiple choices, and microprocessors.

The ever-increasing burden of security, authentication, and identification

The terrorists are coming, the terrorists are coming! And yes, they really are. And not just terrorists, but crooks, thieves, mischief makers, and the curious. All want access to our records and our lives. Our records are not secure, our means of identifying ourselves are laughably impoverished, and the distinctions between security, identification, and authentication are poorly understood, least of all by those implementing the systems that control our lives. Worse, there is a tradeoff between ease of use and security requirements, so that in an effort to ensure perfect security (an impossibility), the security demands are onerous. But there is a paradox: the more thorough the demands of security, the less secure the result. Why? Because when the demands of security get in the way of doing our jobs, we find ways around them. We write passwords on paper, hiding them in insecure locations. We prop open doors, make copies of sensitive material - all because we are dedicated to getting the job done. Thus, the honest workers can undermine the entire security apparatus. Here is a field in desperate need of sanity, one based not only on technical considerations but psychological and social ones as well.

The ever-increasing use of automation

One solution to the problems of complexity and security is to add more layers of technology. Alas, this often solves one set of problems while introducing others. Difficulties with automation have been documented in every domain that has been automated, yet each new domain fails to learn from others but must learn the problems anew. The next domain of automation is the home, the automobile, and medical facilities. Automation can be valuable. If properly deployed it can reduce stress and workload, decrease error and accidents. But if badly deployed it can do just the opposite: increase stress and workload, change the type of error and accident, often to much larger extent than was possible in the non-automated state.

This is what happened in aviation and commercial industrial plants. It is now beginning to happen in medicine, the home, and in automobiles.

The lesson for the field of human-computer interaction

Let us learn from the introduction of technology to other domains. Let us work to ensure that the trend toward higher complexity, onerous security demands, and over-automation is stopped. Mind you, the problems faced in these domains are real, so in reducing complexity, we must still figure out how to give people the choices they want and need. In reducing the complexity of security, we must still manage to increase the security, distinguishing when it is necessary to identify someone from when it is simply necessary to know if they are authorized to use a service. And in introducing automation into everyday things, let us provide the benefits but be ever wary of introducing the perils of over-automation.

Is this list complete? Of course not. I have left off some of my favorite peeves:

Health care. Medical errors. Home care.
Packaging in General: Blister packages in specific

There is lots for us to do.

Don Norman wears many hats, including co-founder of the Nielsen Norman group, Professor at Northwestern University, and author. He is now working on a book on automation in the home and automobile. He lives at www.jnd.org.

References

Klauer, S. G., Dingus, T. A., Neale, V. L., Sudweeks, J. D., & Ramsey, D. J. (2006). The Impact of Driver Inattention on Near-Crash/Crash Risk: An Analysis Using the 100-Car Naturalistic Driving Study Data. Retrieved. from http://www-nrd.nhtsa.dot.gov/departments/nrd-13/driver-distraction/PDF/DriverInattention.pdf.

Strayer, D. L., Drews, F. A., & Crouch, D. J. (2006). A Comparison of the Cell Phone Driver and the Drunk Driver. Human Factors: The Journal of the Human Factors and Ergonomics Society, 48(2), 381-391.