Teaching Functional and Logic Programming with a Single Computation Model

by Michael Hanus

Proc. Ninth International Symposium on Programming Languages, Implementations, Logics, and Programs (PLILP'97), Springer LNCS 1292, pp. 335-350, 1997
© Springer-Verlag

Functional and logic programming are often taught in different courses so that students often do not understand the relationships between these declarative programming paradigms. This is mainly due to the different underlying computation models---deterministic reduction and lazy evaluation in functional languages, and non-deterministic search in logic languages. We show in this paper that this need not be the case. Taking into account recent developments in the integration of functional and logic programming, it is possible to teach the ideas of modern functional languages like Haskell and logic programming on the basis of a single computation model. From this point of view, logic programming is considered as an extension of functional programming where ground expressions are extended to contain also free variables. We describe this computation model, the structure of a course based on it, and draw some conclusions from the experiences with such a course.

Preprint (PDF) BibTeX-Entry Online