WHAT WE WANTED VS. WHAT WE GOT
Part of that is Microsoft's fault. They're the kind of corporation that's used to getting their wish, so when they unveiled Kinect at E3 2009 (back then, it was "Project Natal"), they promised features that couldn't possibly be delivered: intelligent AI that would learn and repeat players' names, scanning of real-life objects for use in-game, instantaneous and responsive Minority Report-style interface controls.
Instead we got mind-numbing ports of iPhone games (Fruit Ninja) and interfaces that require players to hold their hands in the air for several seconds to make a selection. It's like waiting for a high five that will never come.
In place of the 1:1 movement that was promised, our bodies are treated quite literally like controllers—jumping in the air is like pressing the A button, in that it makes a character on-screen jump. But the character jumps higher and stays in the air longer, because your jump was nothing more than an input command. This maintains the disconnect between players and games that's always existed, the same divide that Kinect was supposed to connect all along. (Get it?)