Knowledge Organisation


Kim H. Veltman

Syntactic and Semantic Interoperability: New Approaches to Knowledge and the Semantic Web

Initially written for Dublin Core Meeting in 2000 which rejected the article.
Published: The New Review of Information Networking, vol. 7, 2001.
1. Introduction
2. Historical Background
3. Dublin Core
4. Syntactic Interoperability
5. Semantic Interoperability
6. Concept Systems versus Propositional Logic
7. ICT and the Semantic Web.

1. Introduction

At WWW7 (Brisbane, 1997), Tim Berners-Lee outlined his vision of a global reasoning web. At WWW 8 (Toronto, May 1998), he developed this into a vision of a semantic web, where one could search not just for isolated words, but for meaning in the form of logically provable claims. In the past four years this vision has spread with amazing speed. The semantic web has been adopted by the European Commission as one of the important goals of the Sixth Framework Programme. In the United States it has become linked with the Defense Advanced Research Projects Agency (DARPA).

While this quest to achieve a semantic web is new, the quest for meaning in language has a history that is almost as old as language itself. Accordingly this paper opens with a survey of the historical background. The contributions of the Dublin Core are reviewed briefly. To achieve a semantic web requires both syntactic and semantic interoperability. These challenges are outlined. A basic contention of this paper is that semantic interoperability requires much more than a simple agreement concerning the static meaning of a term. Different levels of agreement (local, regional, national and international) are involved and these levels have their own history. Hence, one of the larger challenges is to create new systems of knowledge organization, which identify and connect these different levels.

With respect to meaning or semantics, early twentieth century pioneers such as Wüster were hopeful that it might be sufficient to limit oneself to isolated terms and words without reference to the larger grammatical context: to concept systems rather than to propositional logic. While a fascination with concept systems implicitly dominates many contemporary discussions, this paper suggests why this approach is not sufficient. The final section of this paper explores how an approach using propositional logic could lead to a new approach to universals and particulars. This points to a re-organization of knowledge, and opens the way for a vision of a semantic web with all the historical and cultural richness and complexity of language itself.