Electronic Brains: Stories from the Dawn of the Computer Age

  • Mike Hally
Granta: 2005. 224 pp. £15.99 Published in the United States as In a Fraction of a Second by Joseph Henry Press ($27.95) 0309096308, 1862076634 | ISBN: 0-309-09630-8

Writing a history of events that took place 50 years ago has the advantages that almost all the documentation is available, and that some of the protagonists are alive to be interviewed. Mike Hally has done this for his book about the development of computers in the years just after the Second World War. But the danger of relying too heavily on such interviews is that important topics can be missed if there are no surviving protagonists.

Electronic Brains grew out of four 15-minute programmes that Hally wrote and produced for BBC radio in 2001. As he accumulated far too much material for one hour of broadcasting, this volume contains much more information than the radio series.

Big deal: UNIVAC was among the first stored-program computers. Credit: BETTMANN/CORBIS

When writing about computers, the question arises as to what you mean by ‘computer’. Most people today would agree that this word implies a machine that is general purpose (it can be programmed to solve essentially any solvable problem), electronic, has a stored program that is held and executed in the machine, and digital, being based on discrete rather than continuous technology. Even so, it is perfectly reasonable to discuss, as Hally does, the pre-history of modern computers, when they were often special-purpose, sometimes mechanical, and seldom stored-program. It is nevertheless surprising to find a chapter devoted to MONIAC (Monetary National Income Automatic Computer), an analogue, special-purpose, hydraulic, externally programmed device built in England for economic modelling. But the chapter is fascinating, and I hurried over to the Science Museum in London to view its MONIAC exhibit.

The chapters on some very early computers, UNIVAC (Universal Automatic Computer, Eckert-Mauchly Computer Corporation, USA, 1951), EDSAC (Electronic Delay Storage Automatic Calculator, University of Cambridge, 1949), and LEO (Lyons Electronic Office, an adaptation of EDSAC for business applications, 1951), are competent expositions about three of the first computers to satisfy the definition in the previous paragraph. This material will be familiar to most computer scientists but not to laypeople, who will find much of interest about the development of these computers and the people who developed them.

The chapter on the RAND 409, a computer built by Remington Rand of Rowayton, Connecticut, is another matter. I admit to prior ignorance of this computer, but the startling thing about this fascinating chapter is that neither here nor in the later chapter on IBM is there any mention of the IBM Card Programmed Calculator (CPC). The CPC was contemporary with, and similar to, the 409, sold better, and was much more influential in the later development of computers. The CPC rates a full page in the 1996 book Computer: A History of the Information Machine by Martin Campbell-Kelly and William Aspray (Basic Books), whereas the 409 is not mentioned at all.

Hally's book also contains chapters on the pre-history of computers in the 1930s and 1940s, as well as on early computers in the Soviet Union and Australia.

The final chapter, on the rise and dominance of IBM, is most noteworthy for the remarkable claim that IBM “doesn't figure much” in the early history of computers, “at least until the mid-1960s”. Actually, as Hally later makes clear, by the early 1950s IBM was already abreast of its only real competitor, Remington Rand, which bought the Eckert-Mauchly firm in 1950. By the late 1950s, IBM was already in a class by itself.

Hally is quite accurate about hardware but, perhaps because he is an engineer, less so about software. One blooper: the American journal Mathematical Tables and Other Aids to Computation, published by the National Research Council in the 1940s and 1950s, is referred to as a “Russian journal”. Still, this book contains enough fascinating material to be of interest to both computer experts and non-experts.