The interface elements have been mentioned in a few reviews, and based on experience with the PS3 Eye I worry that it's a case of something that feels cool at first that's actually the less efficient way to do something.
My issue with most gesture based interfaces - and I've experienced them in business settings, too - is that they tend to actually be slower. This means at first it seems novel and fun, but then it starts to dawn on you that you're actually not using the best tool for the job.
I really like the concepts in Kinect, but I am of a mind that from a gameplay perspective they may turn out to be a little more limiting than Move/Wii while as an interface, until the get voice recognition working everywhere properly rather than a few select places (I don't mean geographically I mean all menu levels) I think the reality is that in many cases you'd actually be quicker with a gamepad.
For example the whole wait a second or two for your selection to be recognized is slow. The Eye on PS3 has the same issues in a few games I've tried. 1 second or more to make this selection, then another second here, then another then I'm playing, whereas is I navigate with the controller I can make all the selections in 1 second or less.
Still, it's early days and both Kinect and - perhaps more importantly the SW behind it - is version 1.0 so I was expecting some form of teething issues.
It's interesting how Fox, plus some newspaper reviews I've read are so clearly not critical reviews and come close to simply being adverts for the system. Not saying they are bought, but It's interesting, playing devil's advocate, to note more technical reviews for Kinect tend towards mixed/good whereas less technical reviews are gushing. So perhaps we're wrong and the gushing reviews are more how an average joe will react to Kinect.
Unfortunately I'm far to technically aware, as an ex-programmer, so see it that way. I see both the promise and the warts - just as I do with Move and do with Wii.