Simple inputs, with high fidelity (maintain accurate, fast, one-to-one translations of even minute movements into input) and low numbers of modes (ideally one), remove a large interaction frustration point for users. The questions they’re continually asking themselves is, “How do I express my current desire to the device?”


Let’s define success a bit differently for computing hardware: “how tightly does your solution fit the problem?” The original iPod’s input devices (and later click wheel) are very nearly perfect: they provide linear and accelerated scrolling, selection, and playback controls (play/pause, back/rewind, forward/fast-forward) with one wheel, a center button, and four pushable locations on the wheel.

There were no other inputs to provide to the device, and there was nothing on the device that demanded complex or convoluted actions to bend these simple inputs into useful conveyance of ideas.


More generally: how well-defined is your problem, and how tightly does your solution fill that potential space? Does it run over or under?