1) You'll move your fingers just as if you were typing on a keyboard, and the sensors will give you feedback when you "type", your arm can be in any position, s

1) You'll move your fingers just as if you were typing on a keyboard, and the sensors will give you feedback when you "type", your arm can be in any position, so long as you move your fingers like you were typing on a keyboard it would work. There won't be actual keyboards. accuracy and speed would ideally be similar to a bluetooth keyboard. 2) Dedicated workspaces are fine, you're just not tied to one concept of a workspace. A park bench can be a dedicated work space, so can standing in front of walls, or a seat at a coffee shop. the ergonomics can be adapted accordingly. 3) Multi-tasking is irrelevant here, if I'm just watching a video for example, I can sit on a sofa, get up and walk around after a bit, go to the kitchen and make some food, go to my office and sit on an office chair,etc.. all the while I would be watching the movie. My posture and location is not tied down to the compute device, that would be the ideal goal. We wouldn't need "smart" devices all that much with this approach either, "things" would interface with my I/O peripherals instead of having their own that requires my interaction. E.g.: imagine getting groceries and you'd navigate the isles and get information about products without a single thing in the store changing, you just have your peripherals interact with the environment better.