Interactive Storytelling is establishing itself as a major application of virtual embodied characters. To achieve further progress in the field, some authors have suggested that it was necessary to break the 10-minute barrier for story duration, while preserving story pace. In this context, understanding scalability issues is an essential aspect of the development of future Interactive Storytelling technologies. Scalability can be defined as the production of a richer narrative which follows the scaling-up of the Artificial Intelligence representations for plot structure or characters roles. We have formalised narrative events in terms of "film idioms" which are dynamically recognised as the story is generated. This enabled us to stage a number of experiments in which we modified several determinants of scalability, such as the number of feature characters or the complexity of their roles and recorded subsequent narrative extension, through the number of film idioms generated.
|Title of host publication||Third International Joint Conference on Autonomous Agents and Multiagent Systems|
|Publication status||Published - Jul 2004|
Bibliographical noteAuthor can archive publisher's version/PDF. For full details see http://www.sherpa.ac.uk/romeo/ [Accessed 30/03/2010]
Charles, F., & Cavazza, M. (2004). Exploring Scalability of Character-based Storytelling. In Third International Joint Conference on Autonomous Agents and Multiagent Systems (Vol. 2, pp. 872-879). IEEE. http://doi.ieeecomputersociety.org/10.1109/AAMAS.2004.10128