On Feasibility of Declarative Diagnosis

TL;DR

This paper explores the potential of declarative diagnosis methods for logic programs, addressing their current limitations and demonstrating how they can be effectively integrated into practical debugging processes.

Contribution

It identifies weaknesses in existing declarative diagnosis techniques and proposes solutions to enhance their applicability in real-world logic programming debugging.

Findings

Declarative diagnosis methods can be improved for practical use.

Weaknesses in current declarative diagnosis are identified and addressed.

Proposed solutions make declarative diagnosis more feasible in actual programming.

Abstract

The programming language Prolog makes declarative programming possible, at least to a substantial extent. Programs may be written and reasoned about in terms of their declarative semantics. All the advantages of declarative programming are however lost when it comes to program debugging. This is because the Prolog debugger is based solely on the operational semantics. Declarative methods of diagnosis (i.e. locating errors in programs) exist, but are neglected. This paper discusses their possibly main weaknesses and shows how to overcome them. We argue that useful ways of declarative diagnosis of logic programs exist, and should be usable in actual programming.