Wednesday, February 21, 2007

The Psychology of Everyday Things (Donald Norman)

Ever stared blankly at your car radio, trying to remember which combination of buttons to push to set the time? Had trouble remembering which light switches corresponds to which lights, even in a place you’ve lived in for years? Struggled to answer call waiting on your new NASA-level phone system? (Hi, Mom and Dad!)

Donald Norman wants you to know one important thing: It’s not you. It’s the design.

The Psychology of Everyday Things (also titled The Design of Everyday Things, depending on country and year of publication)* is a classic among designers—it was, among other things, required reading for the original TiVO interface team. It’s full of hilarious examples of bad design from his own house and other places he’s been (a refrigerator temperature that’s nearly impossible to set, a faucet that befuddled much of a visiting conference, a set of doors that seemed to trap a man between them, the fictitious “Masochist’s Coffee Pot” shown on the cover), thoroughly analyzing why some devices seem so hard to use even for relatively simple things, and what designers can do to make them more usable.

One of his bugaboos is the elevation of aesthetics over usability—he writes off many an unusable object with the dismissal “It probably won a prize.” Aesthetics, for example, often dictates that control panels use rows of similar-looking buttons, which leads almost inevitably to mistakenly pushing the wrong button. This can be inconsequential in a clock radio; in an airplane cockpit or nuclear power plant station, it can be disastrous. (One enterprising team at a nuclear power plant took the step of replacing their identical cookie-cutter levers with different beer keg taps so they wouldn’t mix them up.) This isn’t to say that he thinks aesthetics is a negative value (although it occasionally seems that way), but rather that when aesthetics and usability conflict, they must be balanced appropriately. Aesthetics seems to win out far too often.

He offers a number of principles that can help make design more usable, like making operations visible (users should be able to figure out what they can do with an object—if a phone has a hold function, it should have a “hold” button), providing useful feedback (users should be able to tell what they’ve just done, and whether a particular operation was successful), and exploiting constraints (don’t allow incorrect behavior—Ikea generally does a marvelous job of this, sizing different screws so they only fit in the correct holes and often even making sure that the boards only fit together in the correct direction and alignment). He also devotes a lot of time to the problem of user error, identifying different types of common general errors people make and arguing that designers should design for error—in many cases they shouldn’t even think of it as either “error” or “correct action,” but rather as a sort of conversation between user and device, with the user trying to accomplish tasks in different ways and the device providing feedback on what the best way is to do so. Errors should be both nondestructive and reversible. (In the course of a nice analysis of how Scott Adams managed to mistakenly delete 500 moderated blog comments, Bruce Tognazzini, founder of Apple’s Human Interface Group, notes it is never useful for software designers to blame the user.)

Some of his ideas for improvements seem straight out of a charming 1950s-era World of Tomorrow (such as a scanner on microwaves that can read encoded instructions on packaging so you can just scan them in and push “start”), and at other times he seems like something of a classic Old Coot (such as when he captions a figure showing an instruction manual page from the original NES as “The Nintendo Children’s Toy”). But overall this is a thoroughly entertaining and well-thought-out treatise on user-centered design—and next time you’re befuddled by the shower in a hotel room, you can take a moment to cheerfully assess exactly where and how the designers went wrong.

* This took me a while to figure out. I wanted to read Design, but was stymied by the library listing the status of its single copy as ON SEARCH. There was no way to ascertain what this meant or what might be done about it. Meanwhile, I saw that they had six copies of Psychology, and was dismayed that they had all these copies of a book I’d never heard of and no copies of the well-known one I wanted. Norman would have a field day with this.