Authoring is a complex, knowledge-intensive activity which until recently has been performed exclusively by humans. New computer-based techniques have added horsepower rather than intelligence to traditional approaches, and have not addressed their principal limitations, chief of which is the inability to tailor presentations to individual users at run-time. We believe a model of the user is needed to support this kind of run-time determination of form and content. We describe our approach to the acquisition, representation and exploitation of user models: the most plausible user model is the result of an abductive recognition process and it incorporates assumptions about the user which are then used to constrain the design by abduction of the best presentation. Both recognition and design processes are performed at run-time. We describe a prototypical implementation designed to demonstrate these ideas in the domain of video authoring. Our approach to authoring is intended to apply across multiple media; we have demonstrated these ideas with video because authoring in the video medium with traditional approaches inherits and exacerbates the problems from traditional media, and because the popularity of video as an authoring medium continues to grow.
