JANUARY-MARCH 2007 (Vol. 29, No. 1) pp. 96-99
1058-6180/07/$31.00 © 2007 IEEE
Published by the IEEE Computer Society
Published by the IEEE Computer Society
|References and notes|
PDFs Require Adobe Acrobat
Mike Hally, Electronic Brains: Stories from the Dawn of the Computer Age. Granta Books, 2005, 274 pp., £15.99. ISBN 1-86207-663-4.
Mike Hally, an electronics engineer who worked for British Aerospace for 17 years, now has his own company, Pennine Productions, which produces programs for BBC Radio 4. Electronic Brains originated as a series of 15-minute programs with the same name first broadcast on Radio 4 in 2001. Much of the material for the radio programs came from interviews with the surviving members of the original teams who designed and built the early computers.
Electronic Brains consists of nine chapters; the first deals with the Atanasoff Berry Computer and the ENIAC and the last with IBM's origins and relatively late entry into the computer field. The intervening seven chapters describe both well-known early computer developments and those that are probably relatively unknown. Included in the former group are the Univac and EDVAC in the US; EDSAC, Pilot ACE, and LEO in Britain; the CSIRO Mark 1 (later renamed CSIRAC) and SILLIAC computers in Australia; and the MESM, BESM, Strela, and other computers in the Soviet Union.
Two almost unknown computers get a chapter each. One is the RAND 409, a punched-card electronic computer programmed using a plugboard, built by the RAND Corporation in a Connecticut barn. It went into production in 1951 and was the world's first mass-produced electronic computer. The other is the Phillips Hydraulic Economics Computer, or Moniac for Monetary National Income Automatic Computer, which was designed at the London School of Economics. It used mechanical valves and clear plastic tubing to regulate the flow of colored water around a network of pipes and tanks to simulate the flow of money in the economy.
A short prologue surveys early computing including Pascal's machine, Charles Babbage's Difference and Analytical Engines, and the recent reconstruction of the Difference Engine at the Science Museum in London. The book then ends with a short epilogue that discusses the influence of World War II on the electronic computer's development.
The three appendices give a bibliography and a short account of the preservation (or more often lack thereof) of many of the computers Hally discusses; a short explanation of decimal and binary arithmetic and the solution of systems of linear algebraic equations; and a few notes, titled "Technical Bits," on computer hardware.
In my opinion, the book would have been enhanced by a fourth appendix discussing programming, using perhaps a simple simulated one-address decimal computer, with an example of machine-language programming such as the early EDSAC demonstration program of calculating a table of squares and first differences together with a corresponding program in either Fortran or Basic.
I found Electronic Brains a joy to read because of the author's style and the extensive quotes from his many interviews with computer pioneers. I warmly recommend the book to the computer specialist, historian, and general reader.
University of Alberta
David Leavitt, The Man Who Knew Too Much: Alan Turing and the Invention of the Computer, W.W.Norton & Co., 2006, 319 pp., $22.95, ISBN 0-393-05236-2.
Jim Holt, "Code Breaker: The Life and Death of Alan Turing," The New Yorker, 6 February 2006, pp.86–89.
Since it is relatively rare for The New Yorker to publish a review of a book about the history of computing, it seems necessary to recognize the event with a review of both the book and its review.
The book is a short biography of Alan Mathison Turing (1912–1954) in whose honor the Association for Computing Machinery annually awards a prize of $100,000. Leavitt uses selected material about Turing from the splendid and complete biography by Andrew Hodges. 1 He supplements this abbreviated life with several popularized explanations of what Turing did. It becomes evident early that Leavitt's first goal is to praise Turing. Then, by showing him to have been seriously troubled throughout his life because of his homosexuality, to claim that he had not been appropriately honored and rewarded, but instead had been hounded to his early suicide. Leavitt insinuates that the death was a governmental murder, by first suggesting the possibility and then denying it with the words, "for which there is no evidence, at present, anyway." Leavitt draws attention to movies with characters who he says are Turing's parallels like the title characters in The Man in the White Suit and The Man Who Knew Too much, and one of Turing's favorite characters, Snow White, who bit the poisoned apple that was to put her to sleep until awakened by a kiss from her Prince. Leavitt mentions a few homosexual characters from recent fiction to draw parallels with Turing's social life.
The mathematical ability that Leavitt expects from his readers is exemplified by the fact that he considers it necessary to explain complex numbers. This expectation causes him to devote too many pages to elementary explanations of Turing's rather advanced contributions to logic, mathematics, and computing. Some of the explanations are in error. The book shows evidence of carelessness. On a single page, the author manages to misspell the names of both Eckert and Mauchly and to mess up the name of the Moore School. The printer was unable to handle the exponents of the zeta function, and the publisher's errata sheet did not help.
This polemical recital of the problems of homosexual men contributes nothing to the history of computing and, because of its often unjustified claims of Turing's accomplishments, will give unsophisticated readers a slightly exaggerated understanding of this great man.
Jim Holt's review of the book calls attention to Leavitt's intention and remarks on most of his errors. It also presents a more nuanced, short Turing biography to which one might refer computer-fearing friends. My only complaints are that Holt overstates Turing's accomplishments in two ways. First, he says that Turing was "chiefly responsible for breaking the German Enigma code," which was in fact an accomplishment of many people, some of whom were chiefs. Second, Holt gives Turing credit for "creating the blueprint for the modern computer," another piece of work to which Turing contributed a great deal but hardly created the blueprint. It is generally agreed today that the great leap forward that resulted in the modern computer was the provision of an internally stored program as exemplified in the 1945 proposal for EDVAC, a Moore School document produced by a group which did not include Turing. 2
References and notes
Darren Wershler-Henry, The Iron Whim: A Fragmented History of Typewriting, McLelland & Stewart Ltd., 2005, 331 pp., $29.99, ISBN 0–7710–8925–2.
The book jacket describes The Iron Whim as "an intelligent, irreverent, and humorous history that traces the haphazard trajectory of a writing culture and technology ... [and] dissects our nostalgia for a method of communicating that has all but vanished." As the title implies, this book is about typewriting: its influence on writers in particular and society in general. Author Darren Wershler-Henry—an assistant professor of communication studies at Sir Wilfred Laurier University in Waterloo, Canada—surveys the history of the typewriter's development together with many references. (The references, unfortunately, are listed among the 30 pages of endnotes rather than more usefully in a separate section.) Although the discussions of the social implications of typewriting may be neither particularly interesting nor convincing to a reader of the IEEE Annals, they are enlivened by accounts of some well-known personages and their relation to typewriting. These include David Cronenberg and Naked Lunch, Frank and Lillian Gilbreth and Cheaper by the Dozen, Jack Kerouac and On the Road, Elizabeth Mann Borgese (the daughter of Thomas Mann) and her typing dog, Don Marquis and Archy and Mehitabel. Wershler-Henry offers accounts of many others, as well, including Philip K. Dick, August Dvorak, Henry James, Bram Stoker, Alan Turing, and Mark Twain.
In my opinion, The Iron Whim, although containing much of interest for a variety of readers, would have benefited greatly from more terse and disciplined writing, and by more diligent editing.
University of Alberta
JoAnne Yates, Structuring the Information Age: Life Insurance and Technology in the Twentieth Century, Johns Hopkins University Press, 2005, 368 pp., $49.95, ISBN 0–8018–8086–6.
The literature on the history of computing, while growing substantially in scale and scope over the past two decades, continues to have a severe supply side bias—as scholars and other writers have concentrated on the production of technology (hardware and, to a lesser degree, software) and neglected its consumption. This phenomenon has followed from a larger trend in the history of technology, where scholars focus on invention, innovation, and production to a far greater extent than consumption, maintenance, and use. The relative lack of scholarly attention to the demand side of technology in general, and computing/data processing in particular, has been evident with regard to the scientific and business domains, but given the widespread recognition and growing attention to military, government, and university-based computing over the past decade, has been more pronounced on the business side.
Recently, however, some high-quality scholarship has begun to address this deficiency in the business data processing realm. In 2004, James Cortada published the first of his planned trilogy of works examining the uses and impact of digital computers on different industries: The Digital Hand: How Computers Changed the Work of American Manufacturing, Transportation, and Retail Industries (Oxford University Press, 2003). This insightful study, which provides a rich perspective on a number of important sectors and trades, and gives assessments of broader trends in the economy, is a key contribution to the literature (as will be volumes two and three). JoAnne Yates's new book is also a major achievement to advancing our understanding of business data processing and computing technology use in the 20th century, as well as how institutional users helped shape this development.
In Structuring the Information Age, Yates takes a different, but equally successful, approach than Cortada by producing a detailed examination of the use of digital computing, and the pre-history of earlier forms of data processing, in a single industry: life insurance. Based on extensive research of archival material, trade literature, interviews, and other sources, this study is a valuable model for examining how punched card tabulation machines and digital computing influence an industry's development. Simultaneously, Yates focuses on the reciprocal impact of an industry, and its various organizations, in shaping the design and production of computing and software technology. She demonstrates that life insurance held a special place in its need for sophisticated data processing technology both prior to and following the advent of digital computing, and had a substantial role in influencing developments and designs of punched card tabulating machines, digital computers, and software.
Concerning the pre-World War II tabulation machine era, Yates documents how life insurance was the force behind the invention and marketing of listing or printer technology. Although the life insurance business would not have as direct and clear an influence in shaping computing technology's design and scope in the digital era, she illustrates its continuing role and how the incremental adoption favored life insurance firms and some companies and business strategies over others. In an intriguing analysis of the cooperation within the life insurance trade and broader technical associations (especially the Life Office Management Association and the Insurance Accounting and Statistical Association), Yates details the role of this industry's participants in even wider developments, such as the Codasyl committee that led to the development of Cobol.
In recent years a number of historians, scholars, and other writers have emphasized the importance of IBM's sales and service network, its technological and organizational continuity from tabulating machines to digital computing, its lease structure, and its installed base as factors leading to its rapid success in computing from the mid-1950s forward. Yates paints an evocative picture of how IBM understood the business market better than others did, and learned from its customers in providing computing machines that eased both the financial and technical entry of firms into the digital realm (making the transition gentle and somewhat seamless with the less expensive IBM 650 and 1401 computers and setting the stage for the even more lucrative IBM System/360). She indicates how this continuity did not always hold true with other firms, as was the case with New England Mutual Life. This firm's increasing dissatisfaction with its long-term computer supplier Remington Rand/Sperry Rand's lack of responsiveness to its needs eventually led to New England Mutual Life's shift to IBM in the first half of the 1970s.
Yates describes how shifting suppliers, while beneficial in hindsight, was difficult to see at the time and how loyalty, relationships, and "lock-in" presented great hurdles to such a change. She highlights the fact that it was probably only possible for insurance firms to make such changes because of the knowledge, skill, and persistence of key individuals. In the case of New England Mutual Life, Robert Shafto, a former insurance man and Electronic Data Systems manager, who became the head of computing at New England Mutual in the early 1970s, made all the difference. In detailing this episode, Yates illustrates the importance of individual actors as agents of change and, more broadly, her belief in the importance of Anthony Giddens' structuration theory to the history of business and the history of technology—the duality of structure, where structures (routines, practices, and policies) shape actions, but structures only exist and persist as human agents enact them.
In short, Yates's book—a major achievement in the history of computing literature—is also a substantial contribution to the history of insurance, trade organizations, and more broadly to business history, management theory, and the history of technology. A study of depth and rigor, it will appeal to historians, management scholars, economists, and others. The book successfully models how to blend theory and narrative in a way that tacitly informs and educates on important, understudied topics as it also engages and entertains.
Yates definitively analyzes the life insurance industry's important role in structuring (and, in part, being structured by) the evolution and proliferation of tabulation machine and digital computing technology. She demonstrates the tremendous value of exhaustively examining a single industry, and the power this has for gaining perspective on large questions and key historical trends, and the complementariness of this approach to equally important studies on cross industry comparisons (particularly Cortada's Digital Hand). Yates's book offers countless insights on the structuring of information technology, and breaks new ground in demonstrating content, techniques, and the rich rewards of industry-focused examinations on information technology applications. Her case for the significance of the early, continuing importance of life insurance's reciprocal relationship with the tabulation machine and computing industries is richly documented, well-analyzed, and strong. Yates also highlights the importance of studying the relative influence of other industries and organizations—such as government entities and the aerospace industry, for instance—with equal depth and rigor to gain an even fuller understanding of the structuring of the information age.
Charles Babbage Institute