The Science of Computing: Shaping of a Discipline by Matti Tedre (review)

The main motivation behind The Science of Computing is the observation that there seems to be no agreement on what the field of computing is or what it is about. Researchers and practitioners of computing probably only agree to disagree on what the field is about. Indeed, the main drive behind Matti Tedre’s study is to understand why it is and has been so hard to define computing as a discipline.¹ Its short history is full of fierce debates between different computing actors, each with their own strong, often changing views.

This undefined nature is what makes computing a interesting field of study for the sociologist and philosopher of science. Unlike physics with its long history, the computing discipline hasn’t yet settled down to a well-defined paradigm or set of research programs. One of the open questions toward the end of the book is, in fact, whether it ever will. This resonates well with what is perhaps the book’s main message: the academic discipline of computing is, by its very nature, interdisciplinary and therefore it “defies any disciplinary markers” (p. 214).

This double feature of the computing discipline—its breadth and the lack of a clear identity—requires an approach that is as inclusive as possible and nonetheless can identify the main traditions and narratives that contributed to shaping the discipline. To this end, the book is arranged around three interrelated debates that are considered to have contributed to the shaping of the computing field, each of which revolves around one of three intellectual traditions that are considered to be constitutive of the field: mathematics, engineering, and science. These are basically the three classic research cultures of computing as identified by Peter Wegner in the 1970s.² By ordering these debates more or less chronologically, Tedre sketches a general evolution of the discipline of computing. While in its early years the computing field was struggling to earn its place in academia and declare its independence from other more established fields like mathematics, it has now matured into a field that is by some even regarded as the fourth great scientific domain³ or as a field that pervades all other sciences, dethroning mathematics.

The first debate concerns the positioning of the field as a discipline with respect to mathematics. A short introduction into some of the logical foundations of the field is followed by a discussion of the sometimes opposing viewpoints on the mathematical nature of the computing discipline. This is further illustrated by a discussion of the development of computing curricula, which shows a steady decline of the central place of mathematics in exchange for the growing significance of programming and, ultimately, an evolution toward interdisciplinary and specialized curricula. This evolution of the curricula parallels the well-known formal verification debates (Chapter 4)—probably one of the fiercest debates in the formation of the discipline of computing. After this debate reached its highest and perhaps also lowest point in the late 1980s with the discussion following James Fetzer’s paper in the Communications of the ACM,⁴ it became clear that formal verification was not the solution to the problem of program reliability.

The second debate focuses on the developing relation between computing and engineering. The discussion starts by sketching the history of the stored-program computer, which initially developed at universities but soon moved away from traditional research universities to schools of engineering because it had become a tool rather than an object of research. Attention shifted to programming: the development of the first programming languages and the field of automatic programming (compiling) made apparent that software reliability was a major challenge. Tedre shows how the so-called “software crisis” is rooted in these developments, and the remainder of this section focuses on software engineering debates and the changing face of this crisis. He also shows how software engineering has evolved from a field that was initially the prime (and sometimes easy) target of those who preferred to think of...