Tuesday, November 01, 2005

Rethinking compromise

All design involves compromise, of course. But good design minimizes the compromises made. The first generation (1G) iPod used a scroll wheel which actually rotated and had four clickable buttons around it. (View all the iPod generations here.) In the interest of reducing cost, no doubt, and also reduce the probability of failure, the 2G iPod replaced the movable wheel with a wheel shaped touchpad on which one slid one's finger without the "wheel" actually moving. A sleight of hand -- or finger -- retained the illusion of wheeliness. This was a downshift in usability -- nothing can beat the feeling of actual physical movement to map against changes occurring on the screen. But the shift was not too drastic for users to accommodate.

The 3G iPod eliminated the radially arranged buttons for the same reasons and replaced them with four circular buttons arranged in a row under the screen. Usability took a great hit. The buttons provide no audible or tactile feedback, forcing the user to look at the screen to check for responses from the device. A weak "fix" was incorporated wherein the "buttons" lit up to indicate that they had been pressed. A bigger problem was the layout of the buttons. It was difficult for the user to develop and maintain a mental model of the button layout through some process of mapping. The radial arrangement of buttons was ideal - there was no confusion there. Usability took a dive.

Apple probably knew that and worked quietly behind the scenes for a replacement that would both maintain low cost levels as well as improve usability. The Click Wheel debuted on the iPod mini and was eventually incorporated into the 4G iPod. The buttons went back where they belonged -- around the center of the wheel -- except, now, they were located directly below the wheel. Further, the click was perceptible in both an audible and tactile manner. Usability, though not quite up to the level of the 1G iPod is now pretty close to where it used to be.

Fortunately, Apple seems to employ designers who don't sleep well at night if they screw up, and therefore come back to rectify design problems.

No comments: