The Discrete Event Calculus (DEC) is a logical formalism from the AI community that has been applied in many areas, including robotics, linguistics, story understanding, etc. DEC is distinguished by its use of an explicit time parameter. Traditionally DEC has been used statically - a situation is described and questions posed. Answers come through finding a model. We have turned this around and are using DEC as a programming language for time-dependent systems, where DEC's time parameter is linked to the process clock. In games, for example, it is tied to the frame rate. We have used this for conversational characters but believe it makes a general purpose rule engine for real time processes. We will describe DEC, how we are using it as a programming language, and our implementation, as well as pointing towards some interesting future work.
Slides:
Download the slides for this talk in PDF format.
About the speaker:
Matthew Fuchs has a PhD in Computer Science from NYU where he invented Web Services using SGML, the predecessor of XML. He was a researcher at Walt Disney R&D where he worked on hypertext and the development of XML. He was Chief Scientist for XML Technologies at Commerce One, where he invented data binding (automatically converting XML to and from programming objects), as well as working on multicore technologies at Microsoft and cloud computing. Most recently he has been president of Paideia Computing, a startup focusing on conversational characters for interactive media.
Contact information:
Matthew Fuchs
mattfuchs@paideiacomputing.com