Overview of interactions

Q1. Erik Sandewall:

The paper is a preliminary report of a new action language. What do you expect to achieve by introducing this language which has not already been achieved by existing work on expressing causation in other approaches, such as the event calculus or the TAL languages of Doherty et al?

Q1. Vladimir Lifschitz (5.2):

The action language  C  (as well as the McCain/Turner causal logic that it is closely related to) may be of interest because of the simplicity of its syntax and semantics. Although it has only two kinds of propositions -- static and dynamic laws -- the language is quite expressive, as the examples in the paper demonstrate. The semantics of these propositions is defined in terms of simple and well-understood objects-transition diagrams. And the definition of the semantics in terms of classical logic is just a half-page long.

Second, the view of inertia in  C  is different from what we see in most other work on representing actions. Actions take place against the backdrop of the course of nature. In the past, the focus on inertia has emphasized worlds in which the course of nature is static. Our pendulum example illustrates the fact that inertia-like laws can describe the dynamic course of nature, and they can be overriden by effects described in axioms about actions, just like inertia. In the language  C , both inertia and the inertia-like pendulum law are treated syntactically in similar ways -- as dynamic laws that do not include action names.

This on-line debate page is part of a discussion at recent workshop; similar pages are set up for each of the workshop articles. The discussion is organized by the area Reasoning about Actions and Change within the Electronic Transactions on Artificial Intelligence (ETAI).

To contribute, please click [mail to moderator] above and send your question or comment as an E-mail message.