Gesture layer allows detection of gestures and essentially gives the
user a unified way of getting "complex" input.
It doesn't do (by design) complex gestures like "V" or "S", instead it
does simple and fast interaction gestures like: 2-finger
pinch/rotate/lines or CTRL+Mouse_Wheel zoom and etc.
This lets people to write an application/widget regardless of the input
device used and in a nice consistent way.
Although we consider the API "good", it may change completely. I'm
committing it now so people will be able to start using it, and
hopefully give feedback.
Work by Aharon Hillel with my help.
SVN revision: 61118
Tom Hacohen