On Formal Methods Thinking in Computer Science Education

Abstract

Formal Methods (FM) radically improve the quality of the code artefacts they help to produce. They are simple, probably accessible to first-year undergraduate students and certainly to second-year students and beyond. Nevertheless, in many cases, they are not part of a general recommendation for course curricula, i.e., they are not taught — and yet they are valuable.

One reason for this is that teaching “Formal Methods” is often confused with teaching logic and theory. This paper advocates what we call FM thinking: the application of ideas from Formal Methods applied in informal, lightweight, practical and accessible ways. And we will argue here that FM thinking should be part of the recommended curriculum for every Computer Science student. For even students who train only in that “thinking” will become much better programmers. But there will be others who, exposed to those ideas, will be ideally positioned to go further into the more theoretical background: why the techniques work; how they can be automated; and how new ones can be developed. Those students would follow subsequently a specialised, more theoretical stream, including topics such as semantics, logics, verification and proof-automation techniques.