an entry from

Piotr's R&D blog

At today's Chisel meeting, Karen Parker presented her work on extending users' reach on large tabletops. Her approach is to use a pen to point at items far away and "tractor beam" them closer. She uses a (very expensive) Polhemus pen that senses location and orientation in three dimensions, thus allowing her to compute the matching point on the tabletop. Another approach would be to use a laser pointer with an image-recognition system.

However, our DT tabletops have no cameras mounted above them, and we can't afford a Polhemus setup. We could do the "go-go-gadget-arm" trick, where the user's reach is extended by a special gesture, but I'm not convinced this would be intuitive or easy to control. Instead, I propose the "tablecloth" method: let the user grab the surface and "bunch it up" to bring far away items closer. The distortion should be localized to minimize annoyance to collaborators, and should snap back when the object is grabbed and the auxiliary hand released. This would be easy (easier, at least) to implement if the surface was a deformable mesh, like I suggested in Spacewarps.

Makes me want to revisit the Video Bench again...