@Tim

I agree, it's trivial to accurately decide where and how a user gestured based on your descriptions. And with mode changing, a "blind" person could do plenty.