Every morning, I push the STOP button on the handrail of a number 63 bus. It tells the driver I want to get off at the next stop.

I’m very fond of the button. It immediately radiates robustness: chunky yellow plastic on the red handrail. The command, STOP, is written in white capitals on red. There’s a depression to place my thumb into, with the raised pips of a Braille letter “S” to emphasize its intent for the partially sighted. When pushed, the button gives a quarter-inch of travel before stopping, with no trace of springiness; a dull mechanical ting rings out, and the driver pulls over at the next stop.

It’s immediately clear what to do with this button, and what the outcome of pushing it will be. It makes its usage and intent obvious.

This is a good button.

Buttons have been our primary interface with videogames since Spacewar!—chunky plastic in the arcades; soft compactness on the NES gamepad in our homes; unresponsive, virtual buttons on countless iPhone games that still haven’t found a better alternative.

But how much attention do we, as game players, really pay them? There’s an initial unfamiliarity as we learn a new controller, which sometimes goes hand-in-hand with learning the new motor skills it requires. As we reach mastery, though, the pad and interface fade away, as does the barrier between player and game; and we interact naturally, seamlessly with the worlds onscreen.

That is, if we reach that point of mastery. To begin with, the homogeneity of so many gamepads can be difficult for many players: How many buttons are the same size, the same to the touch, distinguished by only a small icon?

Don Norman describes the problem with buttons and switches at length in The Design of Everyday Things:

Designers love rows of identical-looking switches. The switches look good, are easy to mount, are inexpensive to build, and please the aesthetic sensibilities of the viewer. But they make it easy to err.

You could say the same is true of so many game interfaces. It doesn’t help that most modern game controllers are designed to be multipurpose; the gamepad with your Xbox 360 or PlayStation 3 has to support many titles, many different types of game.

How long do developers and console manufacturers think about the buttons they deal with, these days? I fear it’s not long enough. After all, before the controls of the PS3 were finalized, Sony stated that “controller specifics” were “nothing [they] need to be concerned about as developers.”

Surely they should be fundamental to the design of any interaction?

Good buttons—like the button on the number 63 bus—communicate how they want to be used.

For his talk at Lift10, Russell Davies explained that he was tired of attempts to hide the workings behind public speaking, and had looked for ways to make the delivery of a talk more performative. His solution was a big red button, prominently displayed onstage, used to advance his slides.  He explains:

…it works as I’d imagined. It makes you want to thump it. You lift your hand and bang it down. So if you’re stuck in droning mind-meld mode for a moment you’re soon jolted out of it by the movement and physicality. And your audience is reminded that they’re watching a physical being, not just an avatar. The Big Red Button demands to be thumped. How could you not want to hit its domed surface?

Designer Sascha Pohflepp’s camera Buttons takes a picture—but not of what you saw; rather, it shows you a picture taken somewhere else in the world at the moment you pushed its shutter. It’s a camera that can’t see. And yet it needs to be convincing as a camera: in its form factor, in its operation, and in its physicality. And that means it needs a convincing button; in this case, the shutter release from an Agfamatic 901, chunky, red, and round, an interface that immediately communicates cameraness.

It’s no wonder, given the qualities of good buttons—that they’re immediate, readable, clear in intent—that my favorite game controller belongs to the Nintendo GameCube.

The GameCube pad is marvelous. It’s a riot of shape and textures; each control is different to the eye and to the touch. Instead of diversifying its buttons based on individual games, it diversifies them according to their priority. And so the A button, large and green, is clearly the most important button on the pad. The B button, secondary to it, comes in second place. The X and Y buttons are flanged such that the thumb can roll onto them—or perhaps, as Metroid Prime demonstrated so well, be used to push them at the same time as the A button. At the end of the analogue triggers’ throw is a decisive click, used when maximum pressure is applied to them. And, forlorn above the R trigger is the Z button, slightly unsatisfying to push, and which no one ever seemed to know what to do with; forever consigned to bringing up the onscreen map, or something similarly tangential.

It’s individual, full of character, helpful to the novice, and yet ergonomic and clear for the professional (as many a long Resident Evil 4 session will testify). It manages to be a generic tool that solves many of Norman’s complaints about physical interfaces.

The button is the boundary between our actions and a game. And, though they may be based upon the same micro switches from the same supplier, not all buttons are made equal; they live or die by their design.

Like the STOP button on a number 63, the GameCube buttons make their usage and affordance clear, they respond positively to action, and they’re readable by even a novice user. What more can you ask of a button?

Not a lot. These are the things we should ask—as both designers and users—of all our interfaces.

The Game Design of Everyday Things is a recurring column that looks at how game design and game objects have connections and antecedents everywhere you look. Tom Armitage is a game designer at Hide & Seek and can be found at infovore.org. Top photograph by wlodi