There's been tons of research on tabletop displays for a while now (I built a wine-tasting tabletop display for a course project

), so it's cool to see this stuff coming out in production. BUT, that said, the bitter truth of most of the research that's gone on with this stuff is that though the interaction method is kinda cool, no ones actually found any really good application that really take advantage of the display. I laughed when I saw a pic of this in the paper today because, what was the application it had? Photo sharing. Photo-sharing table-top display papers are pretty much an entire genre unto themselves in CS. It's the only half-decent table-top application anyone's come up with. And everyone agrees that a table top display is a pretty expensive replacement for just printing some photos and spreading them out on a table
So yeah, cool technology, no killer app yet. Mind you getting it into the commercial domain rather than the academic will probably hugely increase the number of useful ideas for it...