Parallel world - gestures: nice to use, but difficult to learn

Gestures on an interactive surface are mostly hard to discover and therefore difficult to learn. That is the major drawback which should be resolved for allowing gesture-based input to develop into an easy and widely used input-technique. A well known gesture is in average faster, than moving to a small button. Gestures do not consume place of an interface and you can apply them on the whole surface. Nevertheless this is exactly the point, where the problem starts. As it is pointed out in the following, the main problem in using gesture-based input is this kind of parallel world. As they are not part of the graphical-interface and most of the time, especially in expert-mode invisible, they are very hard to discover. Overall in most cases gestures are not self-explanatory. That is why users are deterred from using an apparently not intuitive interface. To represent the opposite point of view many interactiontechniques have been explored to give feedback, to support learning and training of gestures, and to make the parallel world of gestures the level beneath the visible interface evident.