Don Norman: Designing For People

Nielsen Norman Group

Stupid Smart Stuff: Watches and Automation

Abbreviated version of article posted in LinkedIn: Stupid Smart Stuff  (registration is not required).

Whenever you see something labeled "smart" or "intelligent," be assured that it is actually rather stupid.

Stupid smart watches

I was deep asleep when my watch started vibrating on my wrist. Chrrr, chrrr, chrrrr, chrrrr - it kept going and going, long enough and strong enough to wake me up. It was 4:30 AM.

I looked at the watch (rather groggily): Battery is almost depleted, the symbol said. Turning off.

"Thank you for letting me know," I said (although with somewhat different words), taking the watch off and throwing it into a corner, where it still lies, a week later.  Thank you Sony for your intelligent smart watch.

If the watch was so smart, why didn't it tell me at 9 PM that it was low on energy and that I should put it on the charger overnight. After all, the watch knows when it is getting low on energy - it even has a little display that shows the battery level, except it isn't on the main display - you have to push some buttons and swipe the screen to get to it.  The watch knows its energy level and time of day - if the watch were a person, I would send them back for retraining.

I was testing the watch for a company I work with. I try out a lot of technology this way. This watch, however, was a low point.  Smart watch? It is really stupid. It can tell the time of day, has a nice stopwatch and timer. Also an alarm clock, although the bizarre interface on that clock had me puzzled into I realized that the way to reset it is to reboot the watch.  (Reboot the watch? yeah. We live in exciting times.)

When I tried the app that gives me two time zones, the watch was not intelligent enough to display both by itself. Smart watch? It couldn't even compute the second time zone without asking for help: It had to make a Bluetooth connection to my phone, ask the phone what time it was in the second time zone, and then display the answer. This smart thing couldn't even do simple addition or subtraction.

The Bluetooth connection to the phone seemed to be the watch's main activity: It was continually connecting and disconnecting even when the phone was in my pocket, less than an arm's length away from the watch. And the Sony designers, bless them, thought that it was important to tell me every time the watch connected or lost the connection. As a result my wrist was always being vibrated. A single powerful vibration, whether for connecting or losing the connection.  First of all, why did I need to know? Second, if I really needed to know, wouldn't it have been better to signify whether the connection was being made or broken instead of using the same signal for both?

Stupid smart automation

This kind of stupidity, especially the lack of communication, is common with smart automation. In commercial aviation, the smart automated systems are also stupid. I've studied several incidents in aviation where the lack of communication to the pilot was deadly. Literally.  Something goes wrong with the airplane, but the intelligent, automatic systems compensate. No need to bother the pilots.  But the wrong thing gets worse and worse until the automation reaches the limit of its compensatory abilities. "OK, I give up," it says, and lets the plane start to crash.

"Huh" say the pilots, scrambling to figure out what happened and what they should do about it.  (Usually they succeed because when a plane is 30-40,000 feet up in the air (9-12 KM), there is quite a bit of time - minutes. Moreover, commercial airline pilots are extremely well-trained.

Automation has now entered the automobile. Alas, the automobile industry refuses to learn the lessons from aviation automation. The automobile engineers believe that they have solved the problems: cars will drive by themselves without any incidents. Humans will monitor the driving and if there ever is a problem, they will simply take over. In fact, the requirement for people always to monitor the self-driving automobiles is now incorporated into the law in some locations. 

How silly. The notion that we can have automated or semi-automated cars as long as the driver is watching over them is a dangerous myth.  As soon as the car can maintain its speed and keep a safe distance from car's in front automatically (already true with adaptive cruise control) and maintain its position in the lane properly (already true with lane-keeping systems), drivers will take the opportunity to find their favorite music, to turn to the rear passengers and converse, to read their email, etc.   It is a myth that people can maintain control when they have nothing to do for a long period.  This myth is well understood in the military and in commercial aviation: it has been studied for well-over 50 years in the field of vigilance (a part of psychology and human factors research).  

In the airplane, the pilots are not attending, but when trouble does arise, the extremely well-trained pilots have several minutes to respond.  In the automobile, when trouble arises, the ill-trained drivers will have one or two seconds to respond. Automobile designers - and law makers - have ignored this information.

It is time to for the designers and engineers of this coming automated world and take heed from the lessons learned over the years in the field of Human-Systems Integration, in studies of automation. Lots of excellent scientists working in the research labs of automobile companies know all this. Product people are notorious about ignoring the wisdom of research groups in their same company.  We now have very smart devices, stupidly done. I fear the consequences will be a lot worse than waking people up at 4:30 in the morning.  Pay attention, engineers: pay attention, designers. Pay attention or people will be killed.

Essays

Books