A Comparison of Direct and Indirect Multi-touch Input for Large Surfaces

Multi-touch input on interactive surfaces has matured as a device for bimanual interaction and invoked widespread research interest. We contribute empirical work on direct versus indirect use multi-touch input, comparing direct input on a tabletop display with an indirect condition where the table is used as input surface to a separate, vertically arranged display surface. Users perform significantly better in the direct condition; however our experiments show that this is primarily the case for pointing with comparatively little difference for dragging tasks. We observe that an indirect input arrangement impacts strongly on the users' fluidity and comfort of ‘hovering’ movement over the surface, and suggest investigation of techniques that allow users to rest their hands on the surface as default position for interaction.