I don't see motion recognition as going anywhere. With keyboards, anyone but the visually impaired can hunt and peck away. It might not be fast, but it can be done. With motion detection, you bring in a much larger group of people who are unable to use the product: elderly, infirm, people with the "shakes" caused by everything from disease to coffee, etc. And the VI still can't use it either, since they can't see what they're supposed to be interacting with. If anything, this will be a supplemental input method.
Voice recognition software again is supplemental, for the problems discussed above. You would need comprehensive grammatical and spelling checkers; if you've ever read a paper by someone who uses a spell checker without really looking, you can imagine the disaster if grammar and spelling were automatically corrected with no proofreading by the user. I work part time for a transcription company, and we found that paying people to type out spoken interviews was much more cost efficient than trying to automate the process and then proofread all the finished materials.
So I agree that keyboards and mouses will be around for a while yet. Some of these other input methods might make some headway, but I don't think any of them will be used more than keyboards.
[blue]Never listen to your customers. They were dumb enough to buy your product, so they have no credibility. - Dogbert[/blue]
