Trying to support conceptual modeling (that is, the usually informal thinking that precludes actual problem solving) with formal uses of logic seems an apparent contradiction in terms. Yet this book shows, via numerous real-world examples, that the first-order predicate calculus, when adapted to problem domains and supported by simple specific tools, can be an appropriate medium to convey ideas while brainstorming about ways to address a given problem.
The introductory material is gathered into the first three chapters, the first of which depicts the whole approach in context. Chapter 2 goes through a simple loan process example to give a flavor of the logic notation (which is closely related to Prolog) used throughout the book, while chapter 3 defines its syntax and semantics.
Various phases of the modeling cycle are then studied in the next four chapters, while logic acts as the underlying guideline. First, logic is shown to be equivalent to graphical modeling formalisms that are used, for instance, in the entity-relationship method (no mention of UML is made here). Then the rapid prototyping approach is translated into the logic framework, using intriguing and unusual examples such as ecological modeling or oil production platform shutdown. Generic modeling is addressed in chapter 6, where so-called “skeletons,” akin to design patterns, are shown to be appropriate parameterizable modeling tools. Even the argumentation material used by designers to support various designs can be expressed in this logic framework, as shown in chapter 7 using other unusual examples such as wildlife reserve selection in Australia or inheritance law.
If extensions to classical logic have been already hinted at in the previous chapters, chapter 8 is entirely dedicated to a very important one, namely temporal logic, which addresses the issue of time modeling, a thorny facet of any real-world design. The book concludes after a short general discussion of modeling language design syntax, semantics, and pragmatics.
Two preliminary words of caution. One should not judge this book by its apparently short pagination (slightly more than 200 pages), since most developed examples are only summarized in the text, the fully developed versions being available on the publisher’s Web site. There is a lot of work embedded in this text. Also, even though the title mentions “lightweight” uses of logic, this is not an introductory book on logic, as attested to by the ample references to Herbrand interpretation, ontologies, and nonmonotonic reasoning.
As a whole, this is an interesting book, based on many years of research by the two authors on the practical application of logic to modeling. Although I found some typos, the text remains clear, though at times a bit abstract. An index and some exercises, together with their solutions, are provided. As such, this research monograph should be worthwhile reading for system designers interested in the more formal aspects of their everyday work.
