College and Research Libraries


functory, this document is nevertheless use-
ful in bringing together in one publication 
references to the growing body of literature 
on library use studies and user behavior in-
vestigations. It will be particularly helpful as 
a source document for other researchers be-
ginning work in this area. Since this state-
of-the-art review at the University of 
Sheffield necessarily precedes the Centre's 
own research and testing, further reports in 
this series of Occasional Papers promise a 
useful contribution in an area in which ·we 
still know far too little. 

·In a further attempt to disseminate the 
progress and results of its work, the Centre 
has begun, as of June 1977, distribution of a 
newsletter entitled CRUS News. While this 
first four-page issue concentrates on news of 
the Centre and its own projects, future is-
sues (no frequency prediction is offered) 
promise to serve as a clearinghouse for 
other activities as well as its own. Individu-
als interested in being placed on the dis-
tribution list for the newsletter or in receiv-
ing a copy of Occasional Paper No. 1 should 
contact the Centre at the University of 
Sheffield.-Herbert S. White, Professor and 
Director of the Research Center, Graduate 
Library School, Indiana University, 
Bloomington. 

Allen, Thomas J. Managing the Flow of 
Technology: Technology Transfer and the 
Dissemination of Technological Informa-
tion Within the R&D Organization. Cam-
bridge: MIT Press, 1977. 320p. $20. LC 
76-57670. ISBN 0-262-01048-8. 
This work consists of a series of reports 

on data collected and conclusions drawn by 
the author from 1963 to 1973. Much of its 
content has already appeared in journal arti-
cles. It is thus not surprising that most of it 
will already be known by anyone who has 
been following the literature on the trans-
mission of information over the past decade. 
On the other hand, it does provide a con-
venient compendium, a state-of-the-art re-
view on the subject for newcomers and 
those who wish to refresh their memory on 
the work reported here. 

From his position at the Sloan School of 
Management at the Massachusetts Institute 
of Technology, Allen was able to investigate 
parallel groups of scientists and engineers 

Recent Publications I 139 

who were working on certain problems 
under government research and develop-
ment projects. At first the information-
gathering processes of those in these proj-
ects were measured and compared to the 
quality of their work. Later, the emphasis 
shifted to determining how information 
enters and flows through a research and de-
velopment organization. As might be ex-
pected, a number of conclusions could be 
drawn from the data-some of them agree-
ing with commonly held views and some 
newer and more startling in their implica-
tions. Among them are: 

1. Engineers think differently from scien-
tists. Scientists are especially interested in 
choosing their own problems and look to 
the community of other scientists for evalua-
tion, therefore, the results of their research 
are fully communicated to the entire re-
search community. Engineers, on the other 
hand, choose to work in situations where 
someone else selects the problems on which 
they will work. Nor is prestige and status 
dependent upon other engineers, but upon 
company officials to whom publication 
means giving secrets to competitors. 

2. In science all work up to a point in 
time is recorded in the literature; in 
technology the literature is less cumulative, 
not built on previous literature, and not 
meant to document the end product or es-
tablish priority. 

3. Whereas information in science is 
transferred via the written record, in 
technology it is more often transferred 
through personal contacts. Other sources of 
information in descending order of impor-
tance to technologists are: customers, the 
company's previous research, consultants, 
and vendors. 

4. At different stages in their work, engi-
neers use the published literature dif-
ferently, spending more time with it at the 
beginning of a problem and tapering off 
markedly about one-third of the way 
through a project. In contrast, internal con-
sulting with colleagues in the company has 
the same initial peak of use but then adds 
another surge two-thirds through the proj-
ect. 

5. In order of importance, engineers use 
textbooks, trade journals, privately spon-
sored engineering journals, professional en-



140 I College & Research Libraries • March 1978 

gineering journals, and handbooks, while 
scientists go first to scientific journals and 
last to textbooks. Whatever source they use, 
engineers tend to acquire and use the mate-
rial for themselves, with recourse to the 
library only about half as often. When the 
library is used, engineers prefer to do the 
searching themselves, rather than obtain 
help from the library staff. In the case of 
unpublished reports, engineers' colleagues 
are the largest source of supply. 

6. Throughout the studies, it was demon-
strated that internal communication within 
the companies was of overwhelming impor-
tance for reaching successful conclusions of 
the assigned projects, with the most suc-
cessful projects using more and more di-
verse local communicants; however, outside 
consultants also played an important role. 

7. In most organizations there were a 
small number of key people to whom others 
turned for information, so-called "technolog-
ical gatekeepers." These people read widely 
in both scientific and technological journals 
and had a broad range of contacts both 
within and outside the company. They were 
thus able to translate information into terms 
that were meaningful for their engineer col-
leagues. Networks of such "gatekeepers" 
also existed, through which the "gatekeep-
ers" themselves maintained communication, 
thus increasing their effectiveness to their 
own groups. Once information entered the 
group, it became diffused through internal 
subgroupings. All of this developed spon-
taneously, with no administrative fiat. 

8. Since communication within a 
technological oi'gariization is so important 
for succe-ss, organizations should strive to 
make such communication easy. Propinquity 
of individuals, good office layouts, or the 
removal of office walls anCl substitution of 
open bays, the location of stairs and 
elevators, and traffic patterns all must be 
examined for this purpose. 

From all his study and experimentation, 
the author comes to the general conclusion 
that much more attention should be paid to 
informal, person-to-person communication 
of technological information within organiza-
tional settings than has been done in the 
past, where the focus has been on the tra-
ditional published literature and the 
framework of supporting bibliographic ap-

paratus. The . author would, it is felt, be 
pleased with the many studies now being 
undertaken to describe the various facets of 
organizational communication in differing 
fields, but it is likely he will be somewhat 
disappointed in his hope that commercial R 
& D firms will begin experimentation 
within their organizations on these topics. 
The need of such firms to make a profit 
probably precludes such rearrangements of 
physical and administrative set-ups.-Estelle 
Brodman, Librarian and Professor of Medi-
cal History, Washington University School 
of Medicine, St. Louis, Missouri. 

Birmingham Libraries Co-operative Mecha-
nisation Project (BLCMP). Final Report. 
Written by D. G. R. Buckle, T. French, 
A. R. Hall, G. N. Metcalfe and D. J. 
Wilkins. Compiled by G. N. Metcalfe. 
Final Report to the British Library Re-
search and Development Department on 
Project Number Sl/G/027, January 
1969-March 1975. Birmingham: BLCMP, 
University Library, 1976 176p. £5.00. 
ISBN 0-903154-05-6. 
This is the final report on the activities 

through March 1975 of the Birmingham Li-
braries Co-operative Mechanisation Project 
(BLCMP). The BLCMP began with three 
libraries (the universities of Aston and Bir-
mingham and the Birmingham Public Li-
braries) and added four additional libraries 
(Birmingham Polytechnic, Bradford Univer-
sity, Warwickshire County, and Aalborg 
University in Denmark) by the end of the 
grant period. After an initial cost analysis 
and feasibility study (comparing manual 
cataloging costs with estimated costs of an 
automated system), the BLCMP elected to 
proceed with an automated shared catalog-
ing system. 

The proJect resulted in the design and 
implementation of the batch computer sys-
tem to utilize MARC records and locally 
generated records in MARC format, the 
creation of a union data base accessible to 
participating libraries, and the generation of 
a variety of outputs required by the partici-
pants. In addition, early project work in-
cluded feasibility studies on the usefulness 
of centrally produced bibliographic records; 
the definition of standards for local record 
variations, cataloging practices, filing rules,