W3C | Submissions

Team Comment on Semantic Web Rule Language First-Order Logic (SWRL FOL) Submission

W3C is pleased to receive the Semantic Web Rule Language First-Order Logic (SWRL FOL) Submission from National Research Council of Canada, Network Inference and Stanford University.

This Submission contains specifications for two overlapping languages: SWRL-FOL and FOL RuleML. These languages build on OWL, RuleML, and SWRL (SWRL Team Comment.) This work is a significant input to the upcoming W3C Workshop on Rule Languages for Interoperability.

The first language, SWRL-FOL, extends SWRL by adding the standard logical connectives such as negation and disjunction from first order logic (FOL, also called first-order predicate calculus, FOPC). This allows certain additional ontologies or knowledge bases to be formally expressed, at the cost of some reasoning performance on certain inputs. One potential problem with this work is the lack of a royalty-free licensing committment from Lucent Technologies, the author's employer.

The second language, FOL RuleML, is a sub-language of RuleML which largely overlaps SWRL-FOL (the first language). RuleML is a lattice of rule languages which has been developed by the The Rule Markup Initiative since August 2000.

Although these languages appear usable in their current state, we expect their primary impact will be as stepping stones to a more unified and perhaps simpler rule language or rule language framework. The languages, in particular, represent a serious attempt to balance the wide range of use cases found around the Web along with a practical sense of what can be implemented. We look forward to seeing a unified approach emerging to cover the wide range of applications for rules and logic. We also look forward to the results of practical implementation experience with these relatively new techniques.

In the near term, the next steps will occur at the W3C Workshop on Rule Languages for Interoperability, which will help bring together the larger community potentially interested in this work. Beyond that, if sufficient interest and focus is seen, a Working Group should be chartered to recommend a rule language for the Web, based in part on the techniques seen in this Submission.

Authors: Sandro Hawke