Interactive storytelling engines
Ong, Teong Joo
MetadataShow full item record
Writing a good story requires immense patience, creativity and work from the author, and the practice of writing a story requires a good grasp of the readers' psychology to create suspense and thrills and to merge the readers' world with that of the story. In the digital writing space, authors can still adhere to these rules of thumb while being aware of the disappearance of certain constraints due to the added possibility of narrating in a nonlinear fashion. There are many overlapping approaches to interactive storytelling or authoring, but each of the approaches has its own strengths and weaknesses. The motivation for this research arises from the perceived need for a new hybrid approach that coalesces and extends existing approaches. Since each of the approaches empowers certain aspects of the storytelling and narration process, the result forces a new research direction which eliminates certain weaknesses exhibited by a single approach, due to the synergistic nature of the various approaches. We have developed: 1) a Hybrid Evolutionary-Fuzzy Time-based Interactive (HEFTI) storytellling engine that generates dynamic stories from a set of authored story constructs given by human authors; 2) a set of authoring tools that allow authors to generate the needed story constructs; and, 3) a storytelling environment for them to orchestrate a digital stage play with computer agents and scripts. We have conducted a usability study and system evaluation to evaluate the performance of the engine. Our experiments and usability study have shown that the authoring environment abstracted the complexity of authoring an interactive, dynamic story from the authors with the use of windows-based interfaces to help them visualize various aspects of a story. This reduces the amount of learning and knowledge required to start having the pleasure of authoring dynamic stories. The studies also revealed certain features and tools that may be reflected by authoring tools in the future to automate various aspects of the authoring process so that the authors may spend more time thinking rather than writing (or programming) their stories.