Solving the Database Update Problem Using Linear Logic


Dong-tsan LEE
Department of Computer Science, The University of Western Australia, Nedlands, Western Australia, AUSTRALIA.
dtlee@cs.uwa.edu.au

C. P. TSANG
Department of Computer Science, The University of Western Australia, Nedlands, Western Australia, AUSTRALIA.
tsang@cs.uwa.edu.au


Abstract

Database update has not been under the same intensity of research as other aspects of databases. The main difficulty lies on the state based semantics of the addition and deletion operations. This cannot be expressed with the static semantics of data models. In this paper, we formulate the database update problem in terms of Linear Logic which was introduced by J.-Y. Girard as a resource-sensitive logic. It has aroused considerable interest from theoretical computer scientists because of its unique way of modelling computation. We demonstrate here that it is possible to represent database updates by using state based properties of Linear Logic. This approach centres around the computation of the next database state. One feature in our approach is that it not only provides real updates rather than database revision, but it also allows the users to reason on previous database states. This approach works well with traditional relational models as well as deductive databases. We also show that this formulation will overcome the semantic difficulty of the retract assert operation as well as the AI frame problem.
Conference Home Page