See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/3330785

The first computers-history and architectures [Review]

Article in IEEE Annals of the History of Computing · May 2001

DOI: 10.1109/MAHC.2001.929915 · Source: IEEE Xplore

CITATION READS
1 477

1 author:

Michael Friedewald
Institute of Electrical and Electronics Engineers
201 PUBLICATIONS 2,477 CITATIONS

SEE PROFILE

All content following this page was uploaded by Michael Friedewald on 01 January 2015.

The user has requested enhancement of the downloaded file.

established itself without question in engi- tecture that eventually became canonical. The
neering and military planning, places where contributors include historians as well as engi-
quantification and modeling are straightfor- neers and computer pioneers.
ward. But as long as social problems persist, This mixture of historians, engineers, and
there will be attempts to apply systems engi- pioneers writing books about computing’s his-
neering techniques to their solution. tory, typical in recent years for the historiogra-
Paul Ceruzzi phy of the computer, has both merits and
Smithsonian Institution weaknesses. While it is certainly insightful to
obtain information from those who actually
Paul E. Ceruzzi is curator of aerospace elec- built the first computers, the different types of
tronics and computing at the Smithsonian contributors have different interests, with vary-
Institution’s National Air and Space Museum in ing results for the reader. In his essay, Robert
Washington, D.C. He recently published A Seidel states that
History of Modern Computing (Cambridge, Mass.,
MIT Press). Currently, he is working on a [the] current interest in the simulation, recon-
research project to document the history of sys- struction, and reactivating of early computers
tems engineering firms located in the Tysons reflects an enthusiasm on the part of the practi-
Corner, Virginia, vicinity. tioners. The historian’s interest in artifacts is dif-
ferent from the practitioner’s. The difference
Raúl Rojas and Ulf Hashagen, eds., The First between their perspectives creates tension
Computers—History and Architectures, MIT between the historian’s use of artifacts and the
Press, Cambridge, Mass., 2000, $39.95, 457 practitioner’s reconstruction of them that should
pp., ISBN 0-262-1817-5. be reconciled if both are to profit from such
reconstructions. (p. 33)
The First Computers contains all 24 papers
that were originally presented at the First Since this tension is also a characteristic of
International Conference on the History of this book, how can the historian of technolo-
Computing, in Paderborn, Germany, August gy profit from it? Almost all pioneers and com-
1998 (see Michael Williams’ report in IEEE puter scientists among the authors describe
Annals of the History of Computing, vol. 21, no. architectural and engineering characteristics of
1, 1999, pp. 72-73). The conference organizers “their” machines in painstaking detail. Their
stated, in the preface, that “we have witnessed essays might be helpful to understand the
a renewed interest in exploring the roots of inner mechanism and the engineering prob-
modern computer technology in recent years.” lems in the construction of the first digital
The organizers could make such a statement computers but provide little help in answering
because machines like Zuse’s Z1 and Z3, the “big questions” in the history of technolo-
Atanasoff’s ABC, the Colossus at Bletchley Park, gy that George Daniels asked in 1970. Raúl
and the University of Manchester’s Mark I have Rojas, for example, states that the internal
been or are in the process of being reconstruct- structure of the Z3, which we would call
ed. They can be reconstructed because enough microcode today, also permitted conditional
material has accumulated over the past decades branching and could therefore “emulate any
to enable a historical comparison of the world’s modern computer” (p. 237). However, Zuse’s
first computer architectures. The papers select- conception of the Z3 was that of a calculator
ed by the program committee dealt with the controlled by a fixed sequence of instructions.
architecture and reconstruction of the oldest Thus Rojas focuses on a possibility—condi-
computers and the contribution of reconstruc- tional branching—that is of little importance
tions to the scholarly historiography of com- in evaluating the importance of Zuse’s machine
puting technology. in its historical context.
Thus the book focuses not on chronology Nevertheless, a number of essays present
(what came first and who deserves credit for it) stimulating new ideas about a unique histori-
but on the actual architectures of the first ography of computing. The essays by Michael
machines that made electronic computing a Mahoney, Paul Ceruzzi, and William Aspray
practical reality. The book covers computers especially construct an insightful framework
built in the United States, Germany, England, for the intellectual and the engineering side of
and Japan. It seeks to demonstrate that similar computing history. Mahoney shows the inter-
concepts were often pursued simultaneously relationship of computer development with
and that the early researchers explored many mathematics and philosophy that resulted in
architectures beyond the von Neumann archi- the emergence of theoretical computer science

April–June 2001 75
 Reviews

designed and built, and analyzes how these

Briefly Noted facilities benefitted from this head start. Using
Dieter von Jezierski, Slide Rules: A Journey Through Three concepts that were originally developed by
Centuries, translated by Rodger Shepherd, Astragal Press, business historians, Aspray shows how and
Mendham, N.J., 2000, $23.50, 136 pp., (paperback), ISBN why IAS and Princeton University—like the
1-879335-94-8. Moore School in Philadelphia and Harvard
Dieter von Jezierski, a long-time employee of Faber-Castell, first University—failed to establish lasting and sig-
published this book in German in 1997. It has been translated into nificant computer research activities.
English by the American physician and slide rule afficionado Rodger Only a few of the essays in this book address
Shepherd. Particularly noteworthy are the discussion of the mate- the issue of reconstructing historical comput-
rials and methods of producing 20th-century rules and the listing of ers, which was meant to be the conference’s
some 22 relatively recent makers and vendors. The focus is on lin- original focus. Besides essays emerging from
ear slide rules, with some discussion of cylindrical models. major reconstruction projects (Manchester
Collectors will find that this book nicely supplements Cajori’s History Mark I, Atanasoff-Berry Computer, Colossus,
of the Logarithmic Slide Rule and the book by Hopp mentioned next. and ENIAC), only the contributions of Robert
W. Seidel and Martin Campbell-Kelly explore
Peter M. Hopp, Slide Rules: Their History, Models and Makers, the value of reconstructing historical artifacts.
Astragal Press, Mendham, N.J., 1999, $35.00, v + 310 pp., Using the examples of the Antikythera mecha-
(paperback), ISBN 1-879-33586-7. nism and the Babbage engine, Seidel shows
In this ambitious volume, Peter M. Hopp has accumulated a that the most significant historical results come
wealth of information about slide rules. Collectors will especially from the contextual interpretation of recon-
appreciate his detailed listing of instrument makers who produced structed artifacts that “illustrate the develop-
slide rules from 1620 to 1900 and of manufacturers’ and vendors’ ment and application of material culture
products from 1901 to 1998. Hopp considers not only linear slide interpretation in a historical context” (p. 45).
rules but various circular and cylindrical models. The book’s interna- Campbell-Kelly in his essay about the soft-
tional scope is particularly noteworthy, as Hopp lists makers from ware simulator of the EDSAC computer argues
Great Britain, continental Europe, and the United States as well as that a physical reconstruction of historical
from Australia, Japan, and China. There is an extensive bibliography. computers is not always necessary to form an
impression of what programming a first-
Stephen Hoffman, “The Navajo Code Talkers: A generation computer was like. He shows that
Cryptologic and Linguistic Perspective,” Cryptologia, Oct. software simulation is a good way to get hands-
2000, vol. 24, pp. 289-320. on experience of early programming tech-
When discussing cryptography in World War II, historians of niques and to validate historical software
computing generally focus on the development of machines for artifacts. In so doing, Campbell-Kelly presents
encrypting and decrypting messages. Undoubtedly these efforts an approach to the historiography of software
had an important role in the history of the electronic computer. At that is worth exploring further.
the same time, the advent of telephone and radio communication The editors state that this volume is of more
encouraged other means of disguising messages. This article than simply historical interest, as architectures
describes the code developed by a group of Navajo Indians and designed to solve specific problems in the past
used widely by them in battlefield communications of the US may suggest new approaches to similar problems
Marine Corps while fighting in the Pacific. in today’s machines. It is, however, a bit too sim-
plistic to expect that historical case studies will
be the blueprints for solving today’s problems.
as an independent basic theory. In doing so, he Moreover, for a modern historiography of com-
shows how far the development of scientific puting technology, only part of this book is of
instruments shapes scientific theories—an value because most of the contributions conform
often-ignored aspect in the history of science. to a dated, artifact-oriented approach.
Paul Ceruzzi succeeds in portraying the his-
tory of computer architecture as far from being Michael Friedewald
a static field with “Nothing New Since von Fraunhofer Institute for Systems and
Neumann.” He shows that many actors and Innovation Research, Karlsruhe, Germany
organizations have contributed ideas that have [email protected]
refined von Neumann’s original design over the
past 55 years. Michael Friedewald is a scientist and con-
William Aspray revisits the Institute of sultant for information and communication
Advanced Study (IAS) and Princeton technologies, innovation research, and tech-
University, where the most influential of the nology assessment. His publications include
first computers—the IAS machine—had been Der Computer als Werkzeug und Medium: Die

76 IEEE Annals of the History of Computing

View publication stats