BKSLVMCH.RVW   980821

"Slaves of the Machine", Gregory J. E. Rawlins, 1998, 0-262-68102-1,
U$12.50
%A   Gregory J. E. Rawlins
%C   55 Hayward Street, Cambridge, MA   02142-1399
%D   1998
%G   0-262-68102-1
%I   MIT Press
%O   U$12.50 800-356-0343 manak@mit.edu www-mitpress.mit.edu
%P   135 p.
%T   "Slaves of the Machine"

What we have here seems to be a fairly erudite version of the Time-
Life guide to computers, only without the pictures.  As such, the
reader could be forgiven for wondering what the title has to do with
the book.  The explanation appears to lie in Rawlins' persistent
insistence on the danger of computers.  While potentially thought-
provoking, the lack of backup and analysis makes it merely annoying.

Chapter one attempts to define what a computer is.  It outlines some
of the work of Babbage and Turing.  (Ironically, although a major
thesis of this section is the idea of the universal tool nature of
computers, the description of the universality of Turing machines is
rather weak.)  Ultimately, however, the flailing words and ideas fail
to portray either an accurate or a useful picture: the bald assertion
that computers manipulate ideas might was well stand alone.  The
description of the inner workings of computers, in chapter two, is
somewhat better.  Somewhat, but not entirely: a current is a stream of
electrons only within the confines of a cathode ray tube, and Rawlins
explanation of the progress from switches to computers stops just when
it is getting interesting (and important).  And, having extended
Moore's Law far beyond its intended life, he ends by telling us that
computers are going to take over, for no particular reason.  There
appears to be an attempt to explain programs in chapter three.  While
a few salient points are made about precision and atomicity, whole
concepts such as the importance of sequence and timing are left out. 
Parts of chapter four might be about programming.  Then again, chapter
four might simply be a reprise of miscellaneous ideas from the first
three chapters.  Rawlins must really love paradoxes since chapter five
contains its own.  Starting out by touching on the idea of
computability (but never quite getting there) it finishes by telling
us that future programs are going to be intuitive--which, if it isn't
non-computable is certainly non-programmable at the moment.  Chapter
six finishes off the book by trying to convince us, on rather
religious grounds, that artificial intelligence is our worst
nightmare.

In attempting to teach about computers in general, the book uses some
interesting and unusual analogies, but isn't really terribly clear. 
It isn't very complete either, and anyone who did succeed in following
the tutorial would likely come away with a somewhat skewed idea about
information technology.  The constant harping on the dangers of
technology is eminently forgettable, since it has neither the
evidentiary background of "Computer Related Risks" (cf. BKCMRLRS.RVW)
or the insight and clarity of "Digital Woes" (cf. BKDGTLWO.RVW).

copyright Robert M. Slade, 1998   BKSLVMCH.RVW   980821