The Computer Boys Take Over: Computers, Programmers, and the Politics of Technical Expertise (review)

In The Computer Boys Take Over: Computers, Programmers, and the Politics of Technical Expertise, Nathan Ensmenger examines the early history of computer programmers and "software crises" in order to "reveal the hidden fault lines within a community: points of tension between groups or individuals, differing perceptions of reality or visions for the future, and subtle hierarchies and structures of power relationships" (11). Though the human element of software, namely the programmers themselves, serve as the centerpiece for his investigation, Ensmenger places these historical subjects within the larger context of the social, political, economic, and technical circumstances of their profession in order to illustrate the complex web of interrelationships that contributed to the creation of its standards and practices.

The book begins by presenting some of the earliest perceptions of programming. Despite the recent prominence of software engineers within the field of computing, Ensmenger points out that programming originated from humble beginnings. When computers were in their infancy during the 1940s and early 1950s, programmers were often viewed as little more than semi-skilled laborers subordinate to prominent hardware engineers. Their role was thought to be nearly clerical, and their contributions given little attention. By the late 1950s, however, programmers had begun to take on a privileged status as highly skilled masters of the arcane. The complexity of early computers and the creativity required to make them perform even simple functions gave their operators an artisanal mystique. Within the greater social context of computing this view of programming as art was both a boon and a detriment to its practitioners. It allowed programmers to "lay claim to the autonomy and authority that came with being an artist" (48), allowing them to demand high pay, high status, and assert their indispensability in a manner that had been denied to them under the previous model of software. However, it simultaneously impeded efforts to turn programming into a standardized profession compatible with corporate structures. Identifying a universal set of technical skills required to be a software engineer was an important step in creating viable training and hiring practices for businesses, but the effort was consistently confused by the strange and haphazard state of programming. Without specific standards, businesses relied upon "aptitude tests" and "psychological profiles" aimed at demonstrating "a particular innate characteristic, or set of characteristics, that could be positively correlated with occupational performance" (63). Though often viewed within the industry itself as unreliable, these hiring practices nonetheless reinforced certain stereotypical images of programmers and defined them as being "born, not made" (81).

It was within this climate that a crisis emerged in the 1960s over a growing labor shortage within a software industry in need of increased professionalization. With contemporary programming perceived to be the realm of eccentric genius, business managers and governmental bodies began to explore methods of de-skilling its practice. This drive was especially manifest in the invention and standardization of automatic programming languages, particularly FORTRAN and COBOL. Though these innovations contributed significantly to software as we know it, they were not functionally simple enough to succeed in wresting the control of software from programmers. An alternative solution to the crisis emerged via the formation of Computer Science as an academic discipline, but ultimately this too proved impractical due to its insistence upon esoteric theories rather than functional training for business. The software industry also proved to be internally intractable as business managers continually struggled to provide worthwhile oversight and direction to programming projects. These circumstances created an environment of competing ideas about what programming was and should be that was continually being negotiated in a back-and-forth struggle for power.

Many of these conceptions of programming and its potential professionalization came to a head at the North Atlantic Treaty Organization (NATO) Conference on Software Engineering in 1968. It was at this meeting that the industry coalesced around the idea of "software engineering," situating computer programming within...