Jade -- Java Diagnosis Experiments

Model-based diagnosis is a successful AI technique for diagnosing physical systems but faces new challenges when applied to finding faults in software, i.e., debugging. Previous work has mostly dealt with modeling the behavior of programming languages with exploitable special properties: logic, concurrent, or functional. In this paper we describe the objectives, the current state, and the addressed research issues of the Jade project. The aim of the project is to develope the theory and practice of applying model-based diagnosis to software debugging of object-oriented programs. The Java language has been chosen as a vehicle for the project due to its relatively simple semantics, its ubiquity (if not now, then in the near future), and its imperative and object-oriented characteristics