Over on ignore the code, Lukas Mathis is worried that the new gestural interfaces are a step backwards. With command-line interfaces, we needed to memorize every command; with graphical, mouse-based interfaces we didn’t have to memorize every command. Now, he thinks, we’re back to having to memorize gestural interfaces.

In a way, gestural user interfaces are a step back, a throwback to the command line. Gestures are often not obvious and hard to discover; the user interface doesn’t tell you what you can do with an object. Instead, you have to remember which gestures you can use, the same way you had to remember the commands you could use in a command line interface.

I think Lukas is forgetting that the mouse originally required instruction, too. It was never obvious how to do things on a mouse-based GUI for those of us who came to them without ever having seen them before. Yes, if you moved the mouse the arrow moved on the screen. If you make the unlikely assumption that people are going to recognize immediately that the mouse needs to be moved, then they will also recognize that moving the mouse moves the arrow.

But so what? They still have to click on a menu and see it pop down; if they click on a file icon, they’ll recognize something’s happened—the file icon gets darker. But what does that mean? I remember the first time I saw a Macintosh; after several minutes I was able to make enough connections to be able to move file icons to the trash can. That was it. I needed to read the instructions to learn what else I could do with it.

The GUI was never obvious before instruction; it became obvious after instruction. No memorization was needed, because, once you knew how the trick worked, then it became obvious. We may not notice it nowadays, because today’s new user’s instruction comes from television, parents, and otherwise watching someone else use the device.

Gestural interfaces can and should be the same. It’s not obvious that you can pinch-and-zoom. A lot of people would never figure out how to do it by chance encounter with such an interface. But once described, then it becomes obvious how to pinch-and-zoom. Flicking right or left; again, not obvious before instruction. But it doesn’t require memorization once you know about it. It’s naturally obvious once you know it exists.¹

Here’s an obvious example I ran into recently while looking at used cars. Two of the nicest cars I looked at, the Jaguar XJ8 and the Lincoln Town Car, have analog clocks above the center console. The clock itself looks nicer on the Jaguar, but the clock overall looks nicer on the Town Car. One of the things I noticed immediately about the Jaguar’s clock is that it’s a really nice clock with two ugly rubber buttons next to it for adjusting it forwards and backwards.

Since I’m not in the habit of adjusting clocks on other people’s cars, I didn’t notice that the Town Car was missing those buttons until I ran across the instructions for adjusting the clock in the manual.

It might be more obvious to you, since you’re reading about it in the context of obvious-once-you’re-told, but the Town Car has two buttons for adjusting the clock as well. They’re the art-deco bars on either side of the clock. I never considered that they were functional when seeing them in the car; now that I know they are I will never forget, regardless of what car I buy.

Very few of the useful things that we do on our computers are intuitively obvious that we can do them; we only know that we can do them because they’ve become ingrained due to well-designed systems (hopefully) that have revealed their natural obviousness. What makes an interface obvious is, once we know the basic gestures (whether using a mouse or a touch-pad) they don’t need remembering, and we know how to use those gestures to do what we think we ought to be able to do in similar situations.