Applying reach in direct manipulation user interfaces

Abstract

The HCI community currently faces the problem of making tangible user interfaces actively responsive to their user's current physical context. This paper explores the context of direct manipulation user interfaces for large horizontal interactive displays. Knowledge of users' reach provides direct manipulation user interfaces with a powerful tool for contextualizing and predicting user action. This paper introduces users' reach as a formal way to predict the previously observed phenomena of workspace segmentation and territoriality. By creating models of "reach-ability", reach probability surfaces can be generated which further explain the impact on workspace usage of the shape, height, and position of the workspace. As the presented techniques build on formal qualitative and mathematical models of reach, they lend themselves particularly well to an algorithmic implementation suited to driving complex user interface behaviour. This paper presents the results of an initial user study to determine the accuracy of these predictions and their underlying hypotheses about reaches role in shaping workspace usage.