From the Editor’s Desk

As the historian Michael Mahoney loved to point out—most famously in the pages of this very magazine—the history of software is hard because software is hard. Software has always represented a difficult conceptual and methodological challenge for historians, in part because it is ephemeral, in part because it does not fit neatly into established categories of science or technology.¹ Historians of computing have had to work hard to develop new methods for finding, documenting, analyzing, understanding, and communicating software history. In this issue, Gerardo Con Diaz reveals an important new concept for thinking about software in the context of the history of intellectual property. In the late 1960s, confronted with the novel and conceptually slippery legal status of software—is a software application written language that can be copyrighted, a device that can be patented, or both/neither?—intellectual property lawyers engaged in a strategy known as “embodied software.” By defining their claims in terms of a specific machine that embodied a specific control program, rather than as an abstract and intangible string of computer code, software firms were, for a time at least, able to mobilize patent law effectively in defense of their intellectual property rights. By tracing this development from its origins in the work of Richard Hamming and Benard Holbrook at Bell Laboratories in the late 1940s through its most notable use in the Autoflow patent of the early 1970s, Con Diaz provides us with an important and foundational new tool for thinking about the history of software in the context of social, legal, and political history.

Continuing on the topic of historiographical methods, James Cortada provides the first in a two-part series on the challenge of writing contemporary history. To the lay-reader, it might seem obvious that writing about recent events is less difficult than writing about days long past; after all, in the modern era, there are numerous records (often in multiple media) available and often even live human beings to interview. But as Cortada reminds us, doing good history is about more than having access to sources. It takes time for a field to develop and for historians to develop useful and appropriate concepts and categories. Mirroring Mahoney, Cortada notes the long and hard work associated with developing the history of software. With every moment, the history of computing becomes more complex, more interesting, and more challenging (as well as, of course, simply more). As both a practitioner and historian, Cortada provides a particularly insightful view into the relationship between business and history and suggest ways in which, by embracing historiographical developments in the broader historical disciplines, historians of computing can continue to expand the scope, insight, and relevance of publications such as the Annals.

In a entirely different type of article, Julie Cohn explores the transition from analog to digital methods in the modeling of electrical power distribution networks. I am particularly excited about Cohn’s article because it connects the history of computing to some of the most lively of the contemporary historical disciplines—namely, environmental history and the history of infrastructure. Her analysis focuses particularly on the continuity between analog and digital techniques. This is not a simple story of one technology triumphing over and displacing another. Cohn tells a complex story about the ways in which sophisticated users made decisions about which technologies to adopt (or not), when, and how.

I first met Maarten Bullynck in his role as one of the organizers of the 2013 Conference on the History and Philosophy of Computing (HaPoC). For those of you who are not aware of this relatively new organization, HaPoC has been doing stellar work bringing together historians, philosophers, logicians, mathematicians, and computer scientists into meaningful and productive conversation. Bullnyck’s article in this issue on computing prime numbers on the ENIAC and EDSAC shows how simple “demonstration” programs were used to both validate digital computer technology and differentiate it from the established tools of analog computing. A general-purpose digital computer could not compute primes faster than existing tools and techniques, but the fact that it could do so at all showed that it...