Open Standards and the Digital Age: History, Ideology, and Networks

By Andrew L. Russell. New York: Cambridge UP, 2014. 316 pp. $32.99 (paperback). ISBN 978-1-1076-1204-4.

It is a commonplace that, however necessary, the development of technical standards is a rather dull business at the best of times. Andrew Russell’s Open Standards and the Digital Age is a useful corrective to this view. In reconstructing the evolution of American engineering standards organizations and processes and using these, as well as international standards institutions, to contextualize the development of compatibility standards within telecommunications and computer networking, the author describes a space marked by contingency and drama. While the author strains on occasion to fit events into his analytical framework, the end result is nonetheless illuminating.

That framework is laid out in the initial chapter and revolves around the concepts of ideology, critique, and openness. Ideology, uncontroversially, constitutes a worldview while critique refers to the process of putting an ideology into practice. From a survey of the concept of openness, the author derives an ideology of openness consisting of commitments to participation and due process, a vibrant market economy, and technological and social progress.

The very breadth of openness as a historical concept, though, can’t help but vitiate to some extent its utility as an ideology. While one can point to some common themes, such a sweeping general term is invariably subject to localized meanings and significance to the point at which one might question its value as ideology. Indeed, rather than establishing the primacy of openness, the book can be read as highlighting its contingencies.

Chapters 2 and 3 recount the development of the organizational infrastructure of American technical standardization. This is grounded in the establishment of the American Engineering Standards Committee (AESC) in 1918, a response to the myriad overlapping and conflicting standards that were being promulgated in the nineteenth century by various entities, including industry trade associations and engineering professional societies. The AESC established a coordinating process that was based in neither the market nor centralized control, but rather in what came to be known, almost reverentially, as the consensus principle. The consensus principle manifested itself as steps to ensure equitable stakeholder representation (though who was viewed as a legitimate stakeholder was another matter), collaboration, and due process to produce rationalized technical standards. When, in 1928, the AESC was reconstituted as the American Standards Association (ASA), which today is the American National Standards Institute (ANSI), the consensus principle retained its prominence.

Chapters 4 and 5 focus on the Bell system and, to a lesser extent, IBM as models of standardization via centralized control, as well as on the reactions to this control. As the author points out, though, while standardization was crucial to managing the complexity of the sprawling Bell system, AT&T did not simply dictate technical standards (of which there was a breathtaking variety) to the constituent companies. Rather, technical standards were frequently the product of interactions across the Bell system and their adoption typically was a function of discussion and persuasion. Moreover, the Bell system was hardly a hermetically-sealed institution and so participated in external standards processes that affected its interests. Nevertheless, AT&T ended up on the receiving end of critiques of centralized control, as did IBM at around the same time in the 1950s. One supposed critique observed by the author was the “competitive critique” (145) which took the form of products produced by IBM’s competitors aimed at addressing underserved market segments. The notion that such basic commercial activity constituted a “critique” embodying an ideology, though, seems a bit of a stretch. While a useful framework in those cases in which a substantive ideology is being acted upon, the author’s analytical approach does not add much value in more straightforward circumstances.

Chapters 6 and 7 tackle the convergence of telecommunication and computer networks. As the need for compatibility standards grew, multiple distinct movements took shape. One arose out of the Arpanet and ultimately produced the Transmission Control Protocol/Internet Protocol (TCP/IP). Another was situated within the International Telegraph and Telephone Consultative Committee (CCITT). Yet another, devoted to development of the Open Systems Interconnection (OSI) standard, played out within the International Organization for Standardization (ISO). However, none of these efforts were pursued in isolation and all were marked by considerable political and technical maneuvering and tension both internally and in their interactions with the others (over the relative merits of virtual circuits versus datagrams in particular).

Anyone familiar with computing already knows how the story ends: with the triumph of TCP/IP as the basis for both the Internet and computer networking generally. That triumph is captured in chapter 8. In the process, some now familiar organizations emerged, including the Internet Architecture Board (IAB) and the Internet Engineering Task Force (IETF), which have had to negotiate their own tensions and conflicts. What to make of all this in the context of standards and openness is less clear.

The author calls attention to David Clark’s famed declaration at a July 1992 IETF meeting that “[w]e believe in: rough consensus and running code,” (253) arguing that this marked a return to the kind of consensus-based standardization highlighted in the early chapters. The irony of that particular moment in light of the somewhat autocratic history of much of TCP/IP development, as charted in the book, is not lost on the author. It was, though, apparently lost on most of the throngs cheering Clark’s declaration.

Given that Internet engineering standards are lauded for their openness, the ultimate failure of the OSI effort is also ironic. The author argues that the OSI process was effectively too open: too democratic and too bureaucratic. (OSI was also undermined, in the view of some, by attempting to create a standard from the ground up rather than by rationalizing existing approaches.) This highlights the difficulty of employing openness as an ideology explaining the evolution of technical standards and standardization processes. It appears that these can be too open as well as not open enough. Historically, the trick seems to have been instituting the right degree of openness at the right moment.

Stuart Shapiro

The MITRE Corporation