Blurring the Boundary between Direct & Indirect Input Shared Environments

Xing-Dong Yang (Supervisor: Dr. Boulanger)

Ever since the introduction of the two input modes in the 1960s, the direct and indirect input has been used in isolation of each other. Indirect input (ex. mouse) is widely used in desktop environments, while direct input is mainly used in mobile devices or tabletop systems. As suggested by other researchers, each input mode has its advantages and limits. New novel interaction techniques were introduced to address the drawbacks of the two input modes, e.g. improving the efficiency for the indirect input or resolving the fat finger problem of direct input. Notice that most of the existing work aimed at resolving the issues of one input mode in isolation of the other one. We suggest that the limits of one input mode are indeed the advantages of the other mode. Therefore, the two input modes could well complement each other in a shared input space, such that users can freely choose which mode to use to get good performance. For instance, the direct input is considered to be: inaccurate, and suffers from occlusion and reachability issues. These problems can be easily resolved by using indirect input because a mouse cursor’s pixel-size tip provides sufficient accuracy when interacting with small objects. Additionally, users can clutch the cursor to easily reach a distant destination which is difficult to reach using a finger. On the other hand, direct input can be used in tasks requiring higher efficiency than what the indirect input can provide.

In this thesis, we explore how to integrate direct and indirect input into a shared input space: mixed mode. This way user will not be limited to a particular input mode, and can thus freely choose which input mode to use based on their needs. In addition to resolving the issues of one input mode, an add-on benefit of having a mixed mode is that users can take the advantage of both input modes.