JARDINE, N., and SIBSON, R. (1968b).

The Construction of Hierarchic and Non-hierarchic Classifications, The Computer Journal Vol 9
pp. 373-380.
LYERLY, S. B. (1967). Some Critical Issues and Problems in Cluster Analysis, Conference on Cluster Analysis of Multivariate Data: Final
Report, pp. 1.01-1.05. Lorr, M., and Lyerly, S. B. (eds.). Catholic University of America, Washington, D.C.
SIBSON, R. (1970). A Model for Taxonomy II, Math. Biosci., Vol. 6, pp. 405-430.
SIBSON, R. (1971). Some Observations on a Paper by Lance and Williams, The Computer Journal, Vol. 14, pp. 156-157.
WRIGHT, W. E. (1972). A Formalization of Cluster Analysis, and Gravitational Clustering, Doctoral Dissertation, Washington University,
St. Louis, Mo.

Book reviews
Digital Computer System Principles, by H. Hellerman, 1973; 466 to 8 inclusively. The first three of these establish design conventions,
pages. {McGraw-Hill, £7-60) introduce the hardware programming language and illustrate the
organisation of a machine by reference to thefivecategories of order
Presumably the title of this book is intended to be parsed either as which it obeys, e.g. input/output, normal operate instructions, read
Digital Computers, System Principles or as Digital Computers, operand, store operand, and branch. The control of the data paths
Systems Principles. Unfortunately its contents do not live up to and registers, that is the timing of the machine, is dealt with at

Downloaded from https://academic.oup.com/comjnl/article/17/4/364-a/443844 by guest on 23 December 2024

either title, in this, its second, edition. When Professor Hellerman length in the remaining two chapters. Chapter 7 deals with the basic
originally prepared this book in the mid-sixties, it covered adequately techniques of timing and refers to 'hardwired' techniques whereas
the principles of Computing Systems as they were then known. chapter 8 is devoted entirely to the concepts, and implementation of
Moreover, his use of APL for the formal description of the mechan- the slower but more flexible technique of micro-programming.
isms and algorithms considered represented a pioneering effort as A general introduction in Chapter 1 which also outlines the objec-
real as, for example, Hellerman's work in creating the first machine tives of the book leads to a description of the organisation and pro-
implementation of Iverson's notation as it was then. Unfortunately gramming of a small instructional computer in Chapter 2. Some of
from the point of view of this book, much has changed in the last the concepts introduced in this way, in particular those of logic and
five years. It is a pity that the publishers chose to issue a new edition storage, are given physical form in Chapter 3 which discusses system
at a time when, in this reviewer's eyes, the book is principally of components. Additional information of a relevant nature is provided
historic interest. in Chapters 9 to 13 which discuss basic communication between sub
A computer today derives its character and its user visible structure systems, input/output techniques and the arithmetic operations of
from its software in general and the operating systems in particular. addition, multiplication and division on an integer and floating point
Yet in this 460 page book there are, according to the index, only two basis.
very brief references to operating systems, on pages 119 and 374 The book concludes with brief descriptions of important features
respectively, though this reviewer noticed additional unindexed from large fast machines in Chapter 14 and special purpose systems
references in the early pages of the book. Any graduate or under- in Chapter 15. These range from a variety of storage techniques such
graduate course teaching Computing Systems Principles or Computer as interleaved memory banks, scratchpad systems, virtual memory,
Science must approach the subject in terms of the total system and the push down memory, through small purpose processors for display
user visible characteristics that it displays. This the book completely purposes and digital filtering, to large special purpose machines such
fails to do, despite the claim in the blurb that 'this extremely revised as Star and Illiac IV. Whilst I could wish that more comment on
text seeks to give the student or professional an introductory yet these last items had been made, it is also clear that space could not
deep understanding of the architecture and design of programming permit this and these topics clearly provide scope for future
and equipment structures in modern computer systems'. publication.
In fact the only real solid reference to programming languages and This book is written in a direct and purposeful manner with careful
software is the second chapter which consists primarily of an intro- selection of the material content. I am pleased to recommend it to
duction to APL and the power of that language in many different all those who have a serious interest in understanding computer
situations plus a brief section on compilers. It has been claimed that organisation and design, ranging from the undergraduate student
the present reviewer should count himself amongst the APL addicts. to the professional engineer. The only drawback is its price.
Be that as it may, he may objectively welcome Hellerman's use of D. B. G. EDWARDS (Manchester)
APL as an educational medium. But this does not make up for the
total inadequacy of this book at this time as an educational tool, as
an introduction to computer systems or as a summary of the
principles that may be associated with their design, implementation Modern Data Processing for Management: a basic systems approach,
and use. Whatever its value in the past, today this book can at best by B. A. Hodson, 1973; 318 pages. (Macmillan of Canada,
be seen as recommended reading for a student of 'classical machine Can. $16.50.)
architectures'. It certainly does not present Principles, or Systems
or otherwise. The last three years have seen a flood of books on business data
processing. Most of them have been aimed at the college or univer-
M. M. LEHMAN (London)
sity student and only a few at the general reader or business executive.
Some authors, Mr. Hodson included, have claimed that their book is
Digital Systems: Hardware Organization and Design, by Fredrick J. equally suitable for both types of reader.
Hill and Gerald R. Peterson, 1973; 481 pages. {John Wiley and This author claims that his book is neither too technical to be
Sons Ltd., £900.) understood by managers nor too general to be valuable for computer
professionals. Most professional computer staff will, I believe,
The book is based on the premise that the natural way to describe disagree with this viewpoint. There are, however, several good things
the organisation of a digital computer is to use a high level program- to be said of the book. It is well presented and readily understood.
ming language and the authors explain their reasons for using APL. Unfortunately, though, most of the examples of hardware systems
The ability to design digital systems and implement them speedily are out of date and most of the economic data is irrelevant for the
with confidence depends very much on this sort of capability, and the same reason. The author's ideas on the methodology of systems
authors are to be congratulated on the timeliness and quality of their analysis and design have, however, stood the test of time. Many of
presentation. It conveys their obvious enthusiasm for the subject his forecasts about the future of computing systems have either
and the liberal use of worked examples, and additional problems in been realised or are well on their way to that state.
each chapter encourages the reader to participate in the design It is difficult to be enthusiastic about this book. It is unlikely to sell
process. in large numbers in this country as many better books are available.
The material essential for their purpose is contained in Chapters 4 j . o. JENKINS (London)

364 The Computer Journal