I don’t know if anyone remembers my blog post on Skinput, but now Chris Harrison has put out an official paper describing more of the ideology and technology behind Skinput. It’s now apparent that Skinput is meant to skip over the use of tables and flat surfaces that it seems as though are concurrently in development. Harrison’s point is that a flat surface is not always readily available, whereas we carry our appendages with us wherever we go. After Harrison goes into more of the reasons why Skinput is the next logical step for mobile devices, he begins to talk a bit about the science behind it. He explains that, when a finger taps your arm, it creates longitudinal and tranverse waves which radiate out until they reach the bone, and are eventually picked up by the bone conduction microphone of the Skinput’s sensor. The thing that surprised me the most about Skinput, is that it is surprisingly accurate. The points which they use to collect data are, on average, 87% accurate which, if you think about how often you can get a touch screen to do what you want it to, is excellent. Another thing which surprised me, was that accuracy changed fairly little with BMI (Although for people in the upper 50 percentile, accuracy did drop to a little over 80%). Overall, I thought the paper was thoroughly interesting, and, at a mere ten pages (nine if you don’t count the bibliography at the end) it’s an easy read. I hope that someday I’ll be able to experience Skinput on a product.
The paper can be found here.