Controlling Interaction with Digital Product Memories
SemProM: Foundations of Semantic Product Memories for the Internet of Things, Springer, Heidelberg, (2013)DOI: 10.1007/978-3-642-37377-0_15
Abstract
Creating intelligent interactive applications based on DPMs is a challenging task. The definition of system reactions, independent interface behaviors, and necessary context data requires expert programmers at various representation levels and often results in hard-to-maintain code. We use the Visual SceneMaker authoring tool, a visual authoring approach which was initially designed for the creation of interactive applications with virtual characters and has been extended to serve as a dialog and interaction manager for our DPM interactive applications. In SceneMaker a clean separation of content (scenes) and logic (sceneflow) is enforced. In order to access high-level context information, the SceneMaker tool provides interfaces to the Object Memory Server (OMS) and to knowledge deduction systems, such as the Java Expert System Shell (JESS). The tool also supports concurrency, variable scoping, and interaction history to facilitate modeling of multiple interaction modalities, robust data access, and flexible interruption policies as they occur in intelligent interactive environments with DPMs. Moreover, the version used allows real-time sceneflow visualization and modification at runtime to facilitate rapid prototyping and code maintenance. In the context of the SemProM project we rely on interactive virtual characters in several application setups in order to provide users with a compelling interaction experience with DPMs.
