College and Research Libraries is forcing decisions on the profession that it may not yet be prepared to consider, let alone resolve. This rather slight volume is the result of a two-year inquiry by Bolt, Beranek and Newman, Inc. , sponsored by the Council on Library Resources. The project director, Dr. Licklider, is a psychologist and is extreme- ly well qualified to undertake such a study. During the period of the study, he was suc- cessively at BB & N, the Advanced Re- search Projects Agency of the Defense De- partment, and IBM (an excellent example of today's high-level commuting scientist). The text is divided into two parts: ( 1) Concepts and Problems of Man's Interaction with the Body of Recorded Knowledge; and ( 2) Explorations in the Use of Computers in Information Storage, Organization, and Retrieval. One of the more important con- cepts is the procognitive system, which "substitutes for the book a device that will make it easy to transmit information without transporting material" (p. 6). Such systems "will not only present information to people but also process it for them, fol- lowing procedures they specify, apply, mon- itor, and, if necessary, revise and reapply" (p. 6). To provide these services "a meld of library and computer is evidently required" (p. 6). The objectives of such systems "are to promote and facilitate the acquisition, organization, and use of knowledge" (p. 21). If we substitute "book" for "knowl- edge" here, we have a definition of li- braries . The substitution, however, is sig- nificant. The development of these systems, by the way, will not only affect libraries serving sophisticated users, but will also have a tremendous impact on the whole educational process, particularly . the rela- tionship between libraries and the learning process. The first half of the book (Man's Inter- action with Recorded Knowledge), despite shortcomings, is well worth reading. The important chapter on "Aims, Requirements, Plans, and Criteria" is a tour de force that does not quite come off. It is lucid, stimulat- ing, and a brilliant monologue, but too much is assumed or left unsaid. There are literally hundreds of ideas here that need closer examination and detailed analysis. This reviewer cannot escape the feeling that it is a hasty generalization "off the top Book Reviews I 407 of the head," skimming glibly over critical problems. Licklider has fortunately left himself and the reader a necessary, if not graceful, exit. If the user of the procognitive system finds himself at a total loss and loses track of what he is doing, he can always press but- tons which ask "Where am I?" or "What should I do next?" "Through either of these programs," the author explains, "the user can reach a human librarian" (p. 127). Yes, Virginia, there is a Santa Claus. The second part of the book (Explora- tions in the Use of Computers) is a brief summary of investigations made during the inquiry. They range from a survey of syn- tactic analysis by computer to methods of evaluating retrieval systems. This part, more than the first, suffers from lack of pattern: a string of isolated studies which pick at a whole bagful of problems without thorough analysis of any one. Many funda- mental problems concerning the learning process, cognition, and semantics must be answered before such studies can be inte- grated into even small operational systems. Despite its short-comings, this book is at present the best and most lucid statement of what the library may look like by the end of the century.-Robert S. Taylor, Le- high University. Specialized Information Centers. By Allen Kent with the assistance of John Canter. Washington, D.C.: Spartan Books, 1965. 296p. $9. (65-16172). It is difficult for this reviewer to say just what this book is all about, or rather what it was put together from. Some of the "case studies" in it are taken from a 1962 National Science Foundation publication; the bulk of it consists of "case histories" taken from an- swers to a questionnaire sent out (prob- ably in September of 1963, although the au- thors don't say) to "500 specialized informa- tion centers ... spread throughout North America." Information centers, and presum- ably the recipients of the questionnaire, are defined, for the purposes of this book as "any library or collection of documents which serves more than one or a few peo- ple ." This is one of the standard ways of manu- facturing a nonbook-great gobbets of un- 408 I College & Research Libraries • September 1965 digested excerpts glued together with a bland, and usually nonoffensive, prose. There are, however, unwritten rules of prac- tical ethics and scholarship to be observed even in this shadow world of nonliterature. One identifies and thanks the contributors, if one ever expects them to answer another survey-and to make it possible for the reader to find out more about a topic which interests him without running into a blank wall of noncitation. One tells how many answers were received, so the reader will know whether the book represents a valid statistical study or random culls from an ever-shortening list. The authors' curious passion for other people's anonymity makes the reviewer's job particularily difficult. There are, for exam- ple, seven footnotes in the whole book. Four of these are to the same ( 1962) book by Kent (a much better book than this, by the way) ; the one reference to a journal article neglects to include the author's name (Fred Whaley). The only information cen- ter cited in the index is, naturally, the Rus- sian Viniti. The text does name, by in- advertent inclusion of a caption, and de- vote some five pages to TDCK, the Nether- lands Armed Forces Technical Documenta- tion and Information Center. (And how did Jan Schuller ever get a questionnaire lim- ited to North America?) The only proper names in the index are those of the pres- tigious Jerome Wiesner (based on an in- correct citation) and Congressman Roman C. Pucinski. James Killian, Presidential Sci- ence Advisor before Wiesner, and hence twice removed from the seat of power rates mention in the text but not the index-a delicate touch of protocol, this. Painstaking textual criticism, on a rainy Sunday afternoon, suggests that not more than seventy answers were received to the five hundred questionnaires mailed, and that not more than thirty-five of these re- sponded in any useful detail. This is, pre- sumably, the ••large sampling" of the pref- ace. (By way of comparison, NSF-61-68, "Specialized Science Information Services in the United States," at one time available from the Superintendent of Documents for $1.75, started out with six thousand mail- ings to a list selected from ten thousand organizations. Some four thousand organiza- tions answered; detailed questionnaires were then mailed to one thousand centers, and answers were returned by seven hundred. The final book, admittedly a directory rather than a survey of operations, includes de- tails on 427 organizations.) These seventy- odd answers were then cut apart and glued together into chapters on acquisition, anal- ysis, terminology control and the like. There are at least two problems with this technique, especially if the first thing done with the editorial scissors is to cut off the respondent's name. By definition, only those centers which answer are included, even if their staff consists of "1/10 profes- sional and 1/ 10 secretary." Slightly larger centers, such as the Defense Documenta- tion Center (which at the time the ques- tionnaires were apparently sent out was still called ASTIA, and so referenced) are dis- missed with a casual remark, even though the literature is bulked with reports de- scribing their activities. Then too, the space devoted to a center tends to be proportional to the length of its response, rather than its over-all impor- tance. For instance, the center which re- ceives most space in the text is the Institute of Experimental Medicine and Surgery of the University of Montreal disguised, pas- sim, as: "the specialized information center of an institute"; "medical library of a med- ical school"; "a medical school with par- ticular interest in documentation ... "; "a medical school with overt interests in the field of experimental medicine"; "the ex- perimental medicine division of a medical school"; "a medical school's information center"; "a medical school's specialized in- formation center"; "a medical school whose information center specializes in the docu- mentation of literature pertaining to endo- crinology and stress"; and, I suspect, as a " ·search-ready' (sic) specialized informa- tion center in the field of medicine." Since my agency is the only federal spon- sor of documentation activities at the Uni- versity of Montreal, I am naturally gratified by the space devoted to these activities. Yet the librarian, George Ember, and I agree that both because of its unique use of the Symbolic Shorthand Notation for coding, and the ready availability in Montreal of skilled yet inexpensive coders (to say noth- ing of the complete absence of mechaniza- tion) the University of Montreal might have gotten slightly more than its fair share of space in the book. I am never quite sure whether the Yalu river should stand between the reviewer .of a book and the advertising claims made ·for it. Since naive book-buyers may be · guided by these, it seems only fair to point out that this is not "The first book in the specialized information center field that. ... " Pride of this place is surely occupied by the classic "Centralized Information Ser- vices-Opportunities and Problems," West- ern Reserve University Press and Intersci- ence, 1958 written, oddly enough, by Allen Kent and James Perry. The librarian, though, should derive some small consolation from this book- the same consolation derivable from one of Samuel Johnson's statements: "Commerce," said he, "can't be so difficult. Look at the class of people who succeed in it." Since the book defines an information center as any library or collection of documents, it not only bridges but annihilates "the exist- ing gulf between the librarian and the documentalist."-Harold Wooster, Air Fo-rce Office of Scientific Research. Clinic on Library Applications of Data Processing. Proceedings. 1963. Ed. by Herbert Goldhor. Champaign, Ill.: Dis- tributed by the Illini Union Bookstore, [1964]. vii, 176p. $2. paper. (65-1841). This small volume serves as an excellent travel guide through selected data process- ing installations in libraries across the coun- try. Representing a broad range of libraries by type, all but three of the ten papers presented are case studies of operating sys- tems in various stages of development. As a result the collection is a practical dem- onstration of how mechanization and auto- mation can help rather than an excursion into theoretical advantages as yet untried. As such, the book is particularly valuable to the novice in the field, although a careful comparison of the variant methods of opera- tion will suggest adaptable alternatives to those librarians already in the systems plan- ning stage. Public librarians will be particularly in- terested in Lorin Burns' description of auto- mation in the public libraries of Lake Coun- ty, Indiana, and John Henderson's very Book Reviews I 409 full explanation of the production of the book catalog in the Los Angeles County system. For those librarians in smaller sys- tems who have justifiable qualms about the expense of mechanization, Burns' figures on the annual cost of handling acquisitions, book processing, registration, circulation control, and catalog card production will probably be a pleasant surprise. James Ja- cobs' paper on the possibilities of data proc- essing in school library systems appears to be valid enough, although for the most part it depicts planning rather than current op- eration. Ralph Parker's paper on the evolu- tion of automatic systems at the University of Missouri, on the other hand, represents more than fourteen years of pioneering progress which university librarians just now beginning to develop systems can ad- mire and envy. Special library systems are represented by two papers, Marjorie Grif- fin's history of the trial-and-error method of development at IBM's Advanced Sys- tems Development and Research Library, and Hillis Griffin's description of processing and circulation at the National Reactor Testing Station Technical Library in Idaho. In the final case study Seymour Taine dis- cusses the preparation of the Index Medicus and the Medical Literature Analysis and Retrieval System (MEDLARS) at the Na- tional Library of Medicine. In the three general papers, Burton Ad- kinson discusses trends in the library ap- plication of data processing, Donald Kraft describes Key Word in Context indexing and the selective dissemination of informa- tion, and Louis Schultheiss contributes a brief but useful exposition of How charts as the basic step in systems design. Includ- ed as an appendix is Edward McCormick's "Bibliography on Mechanized Library Proc- esses," which is an excellent starting point for further investigation. Two conclusions are inescapable in con- sidering the contents of these papers. First, it is apparent that with the exception of ·some efforts of limited scope in the special library field, all applications of data process- ing to date have been · in the area of the library's housekeeping operations, technical services, and circulation. The time is ripe for some significant experimentation in the application of machinery to general biblio- < Continued on page 420)