Technology as a separate species

From the International Herald Tribune (Nov 4, 2007) article on how GPS devices are causing problems in small English villages that can't handle the giant trucks their satnavs systems route them through:

...signs seem to be less and less effective as people increasingly rely more on GPS systems and less on maps, common sense, and their own eyes. ... 'Like if the satnav says, "drive into this muddy field," they think, "that 's weird," but they do it anyway." 

This story (and my own experiences with GPS devices) illustrates nicely the basic premise of Donald A. Norman's new book, The Design of Future Things: It's best to leery and cautious when we start letting our machines do our thinking for us. (Or as I like to say, we need to remember that the operative word in artificial intelligence is artificial.)

 

I am always delighted when Norman publishes a new book. I've been a fan since he wrote Things That Make Us Smart. His Design of Everyday Things always makes my list as one of the most influential books I've ever read.

 

In this book, Norman tackles how we should best design ways to communicate with "smart" machines - cars and houses and appliances especially. (There is little said about computer interfaces, per se.) What happens when cars "help" us drive by not letting us get too close to the car ahead of us or not pull in front of another car? Will we be able move traffic more efficiently or might we be forever be trapped in traffic roundabouts by an overly cautious driving program? Do we really want our refrigerators nagging us about our food intake or our houses selecting music according to our perceived mood? He observes:

If machines can be said to have a "voice," theirs is certainly condescending..." and

"...our products are getting smarter, more intelligent, and more demanding, or if you like, bossy."

Norman cautions that we and machines ".. are two different species," each with its own values, outlooks and objectives. If we recognize this, we have a better chance at communicating effectively. He also suggests we confine machine's task to those that are "augmentative" rather than "autonomous" - helping us make good decisions, not making the decisions for us.

He suggests these "design" rules 

Design Rules for Human Designers of "Smart" Machines

1. Provide rich, complex, and natural signals.
2. Be predictable.
3. Provide good conceptual models.
4. Make the output understandable.
5. Provide continual awareness without annoyance.
6. Exploit natural mappings.

Design Rules Developed by Machines to Improve Their Interactions with People

1. Keep things simple.
2. Give people a conceptual model. 
3. Give reasons.
4. Make people think they are in control.
5. Continually reassure.
6. Never label human behavior as "error."

The writing iis readable and thought-provoking as are all of Norman's books. A good holiday gift for the geek who reads in your house.

P.S. I was delighted that Mr. Norman himself responded to a Blue Skunk blog entry where I stated that I could not get his book in Kindle format. (His comment is about 4th from the last.)  Pretty cool that he took the time to comment. It was like hearning from the Elvis of technology!