Pang: There are two balls you've got here. It looks like one of them is rubber-coated....
Sun: I can tell you a lot of stories about these-- this was one of the things I worked on. Sometimes the ball slipped, and we were trying to figure out why. So I ended up-- [gets picture] This is a scanning electron microphotograph of the steel ball's surface. We were trying to understand whether or not after some use the surface was changing. And we brought in used balls to see if the balls were getting less rough, or whether the peaks and valleys were being leveled with fill from bits of Formica from table surfaces. So we were taking an old record turntable and putting pieces on that, and making a little robot arm that would test to see how many miles it would run before it would slip. In the end we had to coat the balls with rubber, and so we had to justify that cost of coating.
Yurchenco: It turned out that the textured surface of the ball picked up crap. It wasn't just from the Formica, it was plastic wearing off the mouse parts themselves. These wear surfaces here, they shed little bits of plastic, and the first thing the ball rolls over and the first thing it does is pick them up!
Sun: [points to underside of mouse] These little feet would begin to deposit themselves on the work surface.
Yurchenco: The things we were looking at using were just basically very low-quality ball bearings, because they were very precise but incredibly cheap. So were looking at ways of texturing the surface to change its coefficient of friction, and other techniques, but it never did pan out, were were never able to use an uncoated ball in the long run. We had to go to a rubber coating on it. They'd work for a while, but eventually start skipping, and just not cut it.