Logic for Component-based Software Development

Kung-Kiu Lau
Department of Computer Science, University of Manchester
Manchester M13 9PL, United Kingdom
kung-kiu@cs.man.ac.uk

Mario Ornaghi
Dipartimento di Scienze dell'Informazione,
Universita' degli studi di Milano
Via Comelico 39/41, 20135 Milano, Italy
ornaghi@dsi.unimi.it

Abstract:
The title of this paper is styled on that of Kowalski's seminal book 
Logic for Problem Solving [Kowalski:79]. This is because in this paper 
we want to discuss how logic can play a crucial part in next-generation 
component-based software development, just as Kowalski showed in
[Kowalski:79] that logic can be used for programming.

Our starting point is the following quote from the Preface of [Kowalski:79]:

``In contrast with conventional computing methodology, which employs
different formalisms for expressing programs, specifications,
databases, queries and integrity constraints, logic provides a single
uniform language for all these tasks.''

with which we whole-heartedly agree. Unfortunately, despite
this potential advantage, it would be fair to say that hitherto 
Logic Programming has not made any impact on Software Engineering. In
fact, it has missed the boat, as far as the latter is concerned.

We believe that for Software Engineering, logical systems stronger
and more expressive than Logic Programs are needed. In this paper
we want to show that full first-order logic can be used as
a basis for developing a declarative (model-theoretic) approach to
Software Engineering, in particular component-based software development.

Currently Software Engineering is moving from object-oriented to 
component-based development (CBD), but it will not succeed, in our
view, unless components have suitable declarative semantics. We
believe that the declarative nature of logic specifications and
programs will give Logic Programming the chance of a second bite at
the cherry, to become a force in CBD, which has been hailed as 
``the Industrial Revolution for IT''!