Some thoughts on the future evolution of object-oriented database concepts

Abstract

In the last decade, the emphasis in database research has shifted from the relational to semantic and object-oriented models. The latter are supposed to enable us to construct application domain descriptions that are more faithful to how humans conceive reality than what was previously possible. Although current OODB’s are superior to value-based models, nevertheless they are still restricted in their modeling power. To understand how to use them properly, and more importantly to know what to ask for in the next generation of systems, requires that we consider not only purely technical implementation-related issues, but also more abstract problems, such as: What are the distinctions, if any, between objects and values? What are the mechanisms for identification of objects? How are abstractions related to, and represented by objects? Answers to these and similar problems are important for the development of computational object-oriented models, hence dealing with them is in the realm of Computer Science, although they are also of general philosophical interest and have been considered in previous centuries by philosophers, such as Leibniz. These questions, with examples and motivation, are discussed in the paper.

Research partially supported by a grant from GIF — The German Israeli Foundation.