This paper presents an approach to the development of a deductive object-oriented database system, describing the key design decisions and their consequences for implementation. The approach is novel, in that it integrates an object-oriented database system manipulated using an imperative programming language (ROCK) with a logic language for expressing queries and methods (ROLL). The integration is made seamless by deriving both the imperative and logic languages from a single formally defined data model, thereby avoiding impedance mismatches when they are integrated. Key words: Deductive Object-Oriented Databases, Deductive Databases, Object-Oriented Databases, Database Programming, Logic Programming. 1. INTRODUCTION The two most prominent of the emerging approaches to database programming build upon the deductive and the object-oriented paradigms. Both of these approaches have significant strengths. Deductive databases (DDBs) have a formal basis in first order logic, support expressi...

. This paper outlines an approach to the development of a deductive object-oriented database system. The approach presented uses a formally defined object-oriented data model as the starting point for the development of a logic query language and an imperative database programming language. These languages can be used independently -- the logic language for expressing queries or defining rule-based applications, the imperative language for manipulating the database -- or together to support a flexible system for data-intensive application development. 1 Introduction This paper presents the architecture of a deductive object-oriented database system (DOOD) currently under development at Heriot-Watt University, and indicates how such an architecture supports the development of large-scale data-intensive applications. The software development life-cycle traditionally involves a sequence of steps which take the development team through problem analysis, high-level design, detailed design...

We report on our experience in designing and prototyping ROLL, a logical query language under which objectoriented databases conforming to the model defined in [17, 18] can be seen as deductive object-oriented databases. The main contribution of ROLL is to demonstrate the possibility of pursuing for the object-oriented case the same research strategy used in the last decade to extend relation databases with deduction. This paper briefly describes the rationale behind our approach to the field and the DOOD project that is underway at Heriot-Watt. After an introduction to ROLL's underlying data model by means of a simple example, the paper focusses on the implementation of ROLL in a prototype which has been built to experiment with both the language and the model. Certain techniques used in the prototype are described, and ROLL is contrasted with some well-known proposals of deductive languages for object-oriented databases. An indication of future work is given with some conclusions. 1...

This paper is based on our experiences in designing, formalizing and implementing the deductive object-oriented database (DOOD) system ROCK & ROLL [5, 7, 13, 14, 15, 16]. Our overall goal has been to bring together the deductive and the object-oriented approaches in a clean, uniform and practical way for database systems. We have achieved this by first developing a first-order theory that captures features of the object-oriented paradigm in a standard logical framework, and then developing a pair of languages, ROCK (Rule-Object Computational Kernel) and ROLL (Rule-Object Logic Language), that cater for the declarative and the imperative styles in database programming. In this position paper, we concentrate on the form taken by the logical characterization of the object-oriented data model (OODM) that underpins these two language components constituting the system as seen by users. Note that although a friendlier, block-structured surface syntax is available to users of ROCK & ROLL (see [5], for instance), in this paper we use logical form for all our examples as this is closer to the central argument being developed.