BKLEOFBC.RVW   980802

"LEO", David Caminer/John Aris/Peter Hermon/Frank Land, 1998,
0-07-009501-9, U$22.95
%A   David Caminer
%A   John Aris
%A   Peter Hermon
%A   Frank Land
%C   300 Water Street, Whitby, Ontario   L1N 9B6
%D   1998
%G   0-07-009501-9
%I   McGraw-Hill Ryerson/Osborne
%O   U$22.95 800-565-5758 fax: 905-430-5020 louisea@McGrawHill.ca
%P   392 p.
%T   "LEO: The Incredible Story of the World's First Business    
Computer"

On Thursday, November 29, 1951, a computer program was run to
calculate the value of baked goods produced for distribution to the J.
Lyons and Company teashops and retail outlets.  Run weekly from that
point on, this was the first time a computer program had been used for
a strictly office related task.  Most unusually, from the perspective
of those who have been fed the standard history of computers, the
machine on which this program ran was not built by the military or a
large government department, but by the catering company itself: LEO,
the Lyons Electronic Office.

North American audiences may wonder why a restaurant chain would try
to place itself on the cutting edge of information technology.  Lyons,
in post-war Britain, held a position somewhat equivalent to that of
the McDonald's fast food empire in the present day, with fewer jokes
about the quality of the food.  As McDonald's is known for its
distribution system, so Lyons had built a foundation of production
automation and advanced business management.  A 1947 trip to the
United States, in fact, resulted in no lessons they could take from
American companies, but did raise interest in the possible uses of the
new "electronic brains."  (An appendix reproduces the trip report from
that visit, with an amazingly detailed primer on the state of the
computer art.)

The first LEO was, in a sense, a victim of its own success.  Jobs were
undertaken for outside enterprises (including calculation of ordnance
data for the military) and there was increased demand from within the
company as well.  A subsequent generation of machines was built, and
LEO Computers Ltd. was created in hopes that sales might be made to
other companies.  These orders did come, and the programming done by
LEO staff resulted in a very early (and successful) form of expert
system for a steel manufacturer.  The company ran computer service
bureaus and installed computers on three continents.  Eventually,
through mergers, both the company and the LEO architecture
disappeared.

The first part of the book is a basic history of the computer and
company.  With spare, readable prose it recounts the outline of people
and actions involved.  While less personal than some other histories
written by the participants, it is also free of the more egregious
forms of chauvinism that tend to characterize them.  As a part of
computer history, and as one that claims a unique position, it is
rather odd that very little mention is made of computer developments
in other companies of the time.  The lack of hard evidence makes it
difficult to accept the infrequent but invariable position that the
LEO was the most cost effective solution in every competition.

Parts two, three, and four rectify the lack of personality by
including reminiscences from early pioneers, later workers, and those
who dealt with sales abroad.  While differing in style, all these
reports are readable and interesting.

(One item to note that is not strictly part of the book is the list of
contributors at the beginning.  As you read through the short
biographies of the people involved in LEO there are a number who went
on to major roles in British, European, and international computing. 
However, of more interest is the wide range of "extra-curricular"
activities and contributions.)

Because of its position in computer history, this book will
undoubtedly be a fixture on the shelves of historians of technology. 
For those with any interest in the subject, it is also a jolly good
read.  However, the text provides more than its simple and inherent
interest.  The LEO was built and programmed by a group with a very
different background from those who commonly created computer
companies, and particularly those from the US.  This unconventional
outlook provided a distinct style that has much to teach those active
in the field today.

copyright Robert M. Slade, 1998   BKLEOFBC.RVW   980802