Everything else, from the various ways to perform head and motion tracking, to the myriad of input devices and controllers, varies wildly. Coming up with a standard cross-section of these features that most platforms support seems like an excellent idea to me.
I don't envy the people doing this, in terms of controllers:
* There's Leap Motion Orion. Plausibly, there is some future tech involving body tracking.
* Valve doesn't know what the final form of the controller is[1], the current controllers are merely one of the better iterations.
* What we don't know.
There seems to be a need for multiple cross-sections. With the current state of flux in the field, I'm expecting controller to be highly vendor-extended.
for controllers you will need the position/orientation information as well - but that's not much for an API to do.
Again, I feel like I'm missing something
I've done a bit of OpenGL, and that I can understand needing standardization b/c you're talking to really complicated non-standard hardware and the graphics pipeline is pretty involved. But with VR... aren't you ultimately just outputting 2 warped video streams? I understand you have to talk to the hardware a little bit (to get eye and wand/controller positions) and u need to know lens parameters, but it seems like a very thin wrapper around the graphics API.
Am I missing something?