College and Research Libraries


value as well as the appendix which gives 
the examination questions from previous 
years. Some pertinent monographs are con-
tained in the bibliography of suggested 
readings, but a heavy emphasis on public 
library titles is evident. Despite the lucid 
style and a few informative chapters, the 
general paucity of descriptive or inter-
pretive information makes this volume in-
appropriate for general library purchase. 
-]ames Foyle, University of Denver. 

Computerized Library Catalogs: Their 
Growth, Cost, and Utility. By J. L. Dol-
by; V. J. Forsyth; and H. L. Resnikoff. 
Cambridge: M.I.T. Press, 1969. 

The principal value of this book is as a 
catalog of considerations relevant to the 
design of mechanized catalog production 
systems. Some research results and some 
suggestions on specific design features are 
presented. The book is particularly recom-
mended to library administrators and li-
brary systems analysts. Computer jargon 
is used only when necessary, and, when 
used, is defined for the nontechnical reader. 

The "growth" in the title is dealt with 
in terms of the fact that libraries tend to 
grow at an exponential rate. Estimating 
the growth rate for individual libraries can 
be difficult because of the unavailability of 
reliable statistical data. A method of using 
imprint dates as a basis for such estimates 
is suggested. In addition, an original meth-
od of predicting the language breakdown 
of future acquisitions is presented. Using 
this method, the authors predict that for-
eign-language materials will constitute a 
constantly increasing percentage of future 
acquisitions of research libraries-a predic-
tion that will be of interest to all library 
administrators. 

A chapter entitled "An Analysis of Cost 
Factors" concentrates on hardware-related 
costs. It includes a particularly lucid sec-
tion on the problems of choosing a pro-
gramming language, and a useful com-
parison of input devices. The claimed po-
tential for cost savings should be viewed 
warily, since it is not clear what costs are 
included in those presented. A brief ap-
pendix to this chapter, surveying some 
linguistic data manipulation languages, will 
probably not interest the nontechnical 
reader. Another chapter, on typography 

Recent Publications I 123 

and format, discusses the important prob-
lem of achieving maximum information 
density on the printed page while main-
taining legibility. 

Among other values of the book are a 
stimulating discussion of publication sched-
ules for book catalogs and supplements, 
and a chapter on automatic error detec-
tion. It is regrettable (but easily explained 
by the paucity of work on the problem) 
that the latter does not concern itself with 
the more general question of automatic 
editing, since a hefty portion of the cost 
of most mechanized cataloging systems is 
attributable to the necessity of human edit-
ing. In backfile conversion projects espe-
cially, it appears that automatic editing 
routines could be devised that would prof-
itably make use of the large amount of 
organization aheady present in catalog 
card data. 

On-line catalogs are not discussed, prob-
ably because, for most libraries, it now is, 
or shortly will be, feasible to use com-
puters to produce human-readable cata-
logs (perhaps in microform), while plac-
ing the catalog on line is a possibility only 
for the more distant future. A more seri-
ous shortcoming is the failure to discuss 
the use of machine-readable catalog rec-
ords acquired from extramural sources. 
There are serious problems to be solved 
before local systems can make effective 
use of such records, but their availability 
will radically affect the costs of mechaniz-
ing catalog production. Nothing in the 
present book is invalidated when external-
ly produced catalog records are consid-
ered, but to the extent that they are 
available, they must be taken into ac-
count in system design.-Kelley L. Cart-
wright, University of California, Berkeley. 

Directory of Library Consultants. Ed. b y 
John Berry III. New York: R. R. Bowker 
Co., 1969. 141p. $10.75. 

It seems to me that this volume will, 
because it is enumerative and not evalu-
ative, serve a very limited purpose. Li-
brarians of large libraries usually know who 
the real experts are for the projects for 
which consultation help is needed. Repre-
sentatives of small libraries probably do 
not know this and they cannot find out