If this is successful it will be the first completely new interaction paradigm since the mouse. Of course so far interactions are mapped down to hover and click because existing software can then run, but the UI tools aren't limited to that at all.
I'd expect that in five years we'll have a completely different sense of interaction with content through Vision (or Vision like products).
(For those who want to quibble -- Touch is somewhat new relative to the mouse, and multi-touch gestures are truly new but very limited. But the design of apps and user interaction hasn't significantly evolved relative to the mouse-bound era.)