College and Research Libraries of rules for the strategy in which both cat- aloger and user are engaged. We cannot say that we know how users prefer items filed in a catalog, whether a certain user really expects identical items to be together or believes that the alphabetic order has, naturally, scattered them. We have always guessed that we are doing what a mythical majority of the users want, but we have never really known how . many of the users even have a preference, let alone know that some variation is possible. One won- ders why the new filer, eager to show his knowledge of the alphabet, does not repre- sent the untrained user. If a rule is hard for him, it is almost certain to be hard for those like him. Would we be playing this game of strat- egy with more skill if we established a rule and stuck to it, regardless of where the card landed? If so, the rule would have to be devised by catalogers themselves, and it would have to ordain, once and for all, the choice of entry, whether main or added, whether subject or series, according to some objective analysis of which signs and sym- bols are going to equal what. We would have to presume that our object is not to give the user the exceptions we guess he wants, but in every case we would have to show him that we are honestly acihering to the rule which we insist that he must learn if he wants to play at all. One wishes that the Filing Rules could be made a standard part of a high-school course in the use of a library. But, actually, as the alternate rules indicate, they are not quite standardized even for the Library of Congress itself,. and, in any case, no high- school student could read that there is usu- ally a reason for a seemingly arbitrary ar- rangement without thinking that the un- reasonable arrangements predominate, and that the word usually is an outright misap- prehension, if not a lie. He is quite free to assume that the aim of the cataloger has been to play a fierce game of hide-and-seek with him , though he has neither the disposition nor the time , nor-to tell the truth, as he might not-does he have the knowledge to win, even many years later when he is work- ing on his dissertation.-Jay E. Daily) Paula K. Lazarus Memorial Library ) National Con- ference of Christians and Jews) New York. Documentation in Action Documentation in Action; Based on 1956 Conference on Documentation at West- ern Reserve University. By Jesse H. Shera, Allen Kent, [and} James W. Perry. New York: Reinhold; London: Chapman & Hall, 1956. The conference recorded in this book (re- ferred to throughout the text as the Con- ference on the Practical Utilization of Re- corded Knowledge-Present and Future) was held on January 16 to 18, 1956. Some 670 persons attended, including librarians, documentalists, scientists, lawyers, and ex- perts on machine computing, operations re- search, information theory, and language. The object of the conference was "the pro- motion of understanding and cooperation among organizations and individuals repre- senting a wide variety of interests, with four particular foci: a. The use of information and its rela- tion to the structure of recorded infor- mation and the patterns of recourse to it. b. The contribution which certain spe- cialized fields of knowledge, e.g., op- erations research, information theory, etc., might make to improving the uti- lization of graphic records. c. The development and improvement of methods, systems, and equipment for the organization and correlation of in- formation. d. The training of personnel." The book is made up of five parts. Part one consists of seven " 'state of the art' chapters" on fields chosen as basic. These were prepared by committees and distrib- uted before the conference. Part two sketches out what machines, systems, education, co- operation, and language study might con- tribute to better documentation. Parts three, four, and five respectively "summarize the panel meetings," report discussions on pos- sible cooperation in documentation of vari- ous subjects, and assess desirable future re- search. This is a review of the book, . not the con- ference, which the reviewer did not attend. The book is disappointing in matter and 340 COLLEGE AND RESEARCH LIBRARIES presentation. Parts one and two, the for- mal contributions, are mostly ponderous and repetitive to an exasperating degree and often verbose. Matters which must have been commonplace to all present, such as the difficulties caused by the rising flood of publications, the need to base decisions on sound information and the inadequacy of language for exact communication are dealt with repeatedly, at elementary level, and length. Some authors beat about a number of bushes before tackling their subjects. Clotted jargon like "the point of discon- tinuance of implementation" for "the time to stop" (p.46) is mercifully rare, but trip- ping over verbiage is less so, e.g. "Consider the possibilities inherent in the projected construction of the Aswan High Dam Proj- ect on the Upper Nile" (p.172). As a result of all this the " 'state of the art' chapters" make 159 pages (over seventy thousand words). One art that might, in the circum- stances, have been fairly fully treated, the established methods of documentation j li- brarianship, gets less than eleven pages. There is a good chapter in this section on education in librarianship by Egan , Focke, Shera, and Tauber, and a glossary, especially useful for computer terms, by Mack and Taylor. The use of recorded knowledge, a difficult theme, is ·well , if rather tediously, covered by Egan and Hen- kle. They, unfortunately, repeat Bradford's dubious statement that only about one-third of useful papers in science are abstracted,' without later comments on it. Of the six chapters of part two, "Pro- grams for the future ," that by Grosch on machine computers is refreshing in style and downright in approach. Part three , "Discussion," which records very little dis- cussion, has sixteen papers and a report of a discussion on education. The papers are short and more to the point than earlier chapters but uneven in quality. Six are on cooperative and centralized processing in various -fields; the four on language and doc- umentation are useful introductions. Three more are on the application of operations research, information theory, and machine computing to documentation. The eight chapters of part four report six meetings on information processing in .JULY 1957 various fields, one on machine translation and a paper on the programs of UNESCO. The single chapter of part five discusses needed research. How is it that a book written by many distinguished people adds so little to our knowledge? It looks as if the conference tried to do too much. "Cooperative infor- mation processing" has failed in many fields; the sections on it deal largely with centralized processing, and the whole could probably have been assigned to a separate conference without loss. Education for li- brarianship has been discussed much and often by those competent to do so. Little good was done by fresh discussions with others. There seems to have been little con- trol or coordination of papers read ; far too much irrelevance got in. It is ironical that so many writers on this subject do not see the importance in "utilization of record a knowledge" of the cle ar and concise record- ing of knowledge. The feeling among schol- ars that short, clear words and sentences are unscholarly dies hard. All this could have been overcome by thorough editing and se- lection, and the book cut by at least a third. As it is, the librarian and documentalist will learn little from it, and the laymen who (judging from the blurb) are expected to read it will probably lack the needed perse- verance.-D. ]. Campbell, Aslib, London. Solving Library Problems: A Comment The article by Fernando Pefialosa and the important announcement of the establish- ment of the Council on Library Resources, Inc., both in the November, 1956, issue of CRL , called to mind a rather puzzling thing about librarians. Why is it that so many suggestions, such as that by Mr. Pefia- losa, are made and so little is done about them? May I offer as an answer that we have no valid way of testing the suggestion in ad- vance? Our only way of dealing with these and other suggested improvements is to re- treat behind the statement that trying it out would cost too much money. This makes me wonder if the Council on Library Re- sources, Inc., will not come to merit the defi- 341