Logic Modelling Workshop

Background and Purpose of the LMW

Research on Reasoning about Actions and Change has a great need for concrete application examples - axiomatizations of scenarios of nontrivial size. Such scenarios are needed for several purpooses:

• To give examples of existing approaches and to show how they work;

• As the basis of comparisons of different existing approaches;

• To identify weaknesses in current approaches, and thereby to give the reasons for extensions of current approaches as well as for entirely new ones;

• To provide test examples for software systems based on logics for actions and change.

By tradition we have been using toy examples, that is, examples that can be defined, discussed, and axiomatized in the course of one or two pages. However, such toy examples are no longer sufficient, and that we need to proceed to larger ones. For this purpose it is necessary to have some method of modularization and successive development: instead of defining the axiomatization of each scenario from scratch, it would be preferable to have a system whereby later axiomatizations can build on earlier ones.

It would also be very useful to have a possibility of common debate about the axiomatizations. It is likely that axiomatizations will sometimes need debugging, and there may also be a need for discussion about the pros and cons of various axiomatizations, within the same choice of logic or between different logic approaches.

Finally, it will be desirable and sometimes possible to have a formal analysis of proposed axiomatizations, for example, for proving them sound and complete with respect to a proposed underlying semantics, that is, a pre-logical but formal account of the intended application domain.

For all of these reasons, the field needs a way of publishing axiomatizations, so that they are stable, publicly available and retrievable, and citable in a reliable way. However, the publication traditions in the field does not recognize an axiomatization as a topic for a research article in a journal or conference. Our traditions tend to favor longer articles, and articles that propose new logics rather than using existing ones.

In order to remedy this situation, the ETAI area of Reasoning about Actions and Change will start an on-line, Internet-based activity called the Logic Modelling Workshop. The basic idea is to define accounts of axiomatizations as technical notes that are published in the official but non-refereed News Journal on Reasoning about Actions and Change. This provides the formal status and the citability, but without confounding with the peer reviewed articles (with traditional criteria) that appear in the ETAI (Electronic Transactions on Artificial Intelligence). At the same time, we set up a webpage structure that provides overview of these axiomatizations and a forum for the discussion about them.

The Electronic Newsletter on Reasoning about Actions and Change will include short abstracts of these technical notes, but not their full text. If the number of discussion contributions concerning the notes becomes large, we may also separate them into a Newsletter of its own, but for the time being they will be included in the existing Newsletter.

Maintenance information:

Latest update 18.6.1999 by EMTEK group. Edit mode aml, position code C.lmw.rules.background.