College and Research Libraries The Indexing of Current Research Information by the Library and Graduate Engineering Students BECAUSE the School of Engineering, Air Force Institute of Te.chnology, estab- lished the first college degree program in reliability engineering, the institute li- brary has been collecting hundreds of documents in this fast-growing field. An immediate problem was to classify and to store the material in such a way that it would readily be available to a faculty member or student interested in locating specific reference documents. The problem of classifying and storing the information was solved through care- ful study by a group of interested librar- ians and faculty members. It was clear from the outset that the Dewey decimal system imposed severe limitations on the flexibility needed. The very nature of re- liability as a new field cutting across the boundaries of many disciplines made ex- tensive cross-referring necessary. If this flexibility was to be made possible at a reasonable cost in effort and resources, an automatic data processing system seemed to be the answer. Finally, a punch card system was decided upon because of the following advantages: (1) reference to specific areas could be culled from the main list by an automatic sorter; (2) ref- erences could be added and placed in the proper space by using automatic equip- ment; (3) references could be duplicated automatically; (4) a complete or partial bibliography could be compiled and printed automatically, and supplements could be regularly issued. The committee then set up a classifi- Bv HARRY E. HAND Dr. Hand is Associate Professor and Head , Department of Humanities, School of En- gineering, Air Force Institute of Technology , Wright-Patt erson Air Force Base, Ohio. cation code consisting of ten rna jor cate- gories, each containing up to ten sub- categories. The committee, of course, tried to establish a classification code which would reflect the needs of those who would have to work most closely with it. Each major category took about four hours to complete. After the classification code had been devised, there remained the task of ac- tually indexing or coding the reliability documents according to the classification code adopted. The library itself quickly prepared standard bibliographical refer- ence cards which included the usual in- formation of author, title, document number (if any), source, and last but not least, the all-important code or classifi- cation number for each document. All of this information eventually was to be arranged so that it could be coded on the punch cards. In addition, the library undertook to write fifty-word descriptive abstracts for all of the material. Conceiv- ably, the abstracts could be included as "follow-on" cards in the automated sys-· tern. Thus a fully annotated bibliography on punch cards would become a reality. At this point, the library proposed the idea of using graduate students in relia- bility engineering to help index the ref- 464 COLLEGE AND RESEARCH LIBRARIES erence material. The library did not have enough time and manpower to cope with the vast number of reliability docu- ments collected on its shelves. Moreover, the members of the library staff did not always have the technical knowledge of reliability engineering to enable them to perform meaningful indexing and ab- stracting. Twenty-five graduate students, members of the 1963 class in reliability engineering, provided a potential source of the knowledge required. Thus they were assigned a special project as a part of English 310-Seminar in the Com- munication of Technical Information, a required two quarter-hour credit course taught by the humanities department. Since one of the purposes of this course is to discuss special problems in the stor- ing and retrieving of technical reference material, the use of student help in cod- ing the reliability documents seemed ap- propriate. Each of the students in the class was given eighty of the standard bibliograph- ical reference cards already prepared by the library. A major code number in keeping with the classification system was to be placed on each card, and more than one major code number could be assigned if cross-referring seemed advis- able. In order to assign valid code num- bers to the cards, the students either had to scan the appropriate reliability docu- ment itself or consult an adequate ab- stract, if one were available. In keeping with the efforts of the library, each stu- dent also was to write a fifty-word de- scriptive abstract. Again, the student either had to scan the document itself or he could base his abstract on one that he might find already printed in a reference source. In any event, the student was re- sponsible for making sure that his de- scriptive statement was both accurate and concise. There was one more important re- quirement for the students. At the end of the term, students were to submit, along with the coded cards and abstracts, NOVEMBER 1963 a five-page analysis and evaluation of the classification code which they had used to number the documents. Specifically, the students were asked to judge the classification system, both from an in- dexer's and a potential user's point of view. Because the National Aeronautics and Space Administration had also de- vised a classification system for filing re- liability information, students were given copies of this system. A few students used both the NASA and the institute classifi- cation systems to index the assigned doc- uments, and then wrote an essay com- paring the two systems. All papers were to be read by the instructor and by the librarians and faculty members involved. The efforts of the students resulted in positive benefits for the classification code, for the students themselves, and for the library. The advantages for the classification code itself have been tan- gible. Although the students were pri- marily concerned with the ease of filing the information, all of them took a criti- cal look at the reliability bibliography from the standpoint of a potential user. The student papers listed several sug- gestions about the classification code which the library staff and the project committee intend to incorporate into the system. A more useable system should re- sult. From the standpoint of classroom in- struction and learning experience, the students themselves realized several im- portant advantages. Most students felt that they did not spend a disproportion- ate amount of time on the assignment, considering the insights which they gained. First of all, students located, and with the help of the library scanned nu- merous publications in the field of relia- bility engineering. In some instances the references suggested possible thesis top- ics, and in most instances the material provided valuable background and bib- liographical references for theses in prog- ress. Second, a know ledge of the kinds of source materials available in reliability 465 engineering would certainly aid the stu- dent in his later professional career. Third, the students learned how to go about writing short abstracts, and for the first time some of them realized both the library. Last, and perhaps most impor- tant, students came to appreciate the im- portance of a good technical library in filing and locating documents. The tools and tasks of librarians took on a new difficulty and ·importance of preparing meaning for the students. helpful abstracts as guides to researchers. The unmi~taka?Ie conclusion is that Fourth, the graduate engineering stu- - graduate enginee:Ing s~ude~ts can I?rofit- dents clearly saw that engineers who need - ably help technical hbranes to mdex to retrieve information from an auto- l~mited amount_s of. technical informa- t d t h ld t k d' t · twn. The classification system used to rna e . syhs e 1 m. s oul'b a e aff Irect In- code the information thus will be closely terest In e ping a I rary sta cons ruct . . b h h 1 h 1 · fi · d f fil' h . scrutiniZed y t ose w ose eventua con-t e c a_ssi catiOn co e or Ing t e In- cern is to retrieve information. A knowl- formatwn. In short, they became aware d f th bl · 1 d · th h e ge o e pro ems 1nvo ve In e t at no automated reference system can classification of information and a famil- be any better than the classification code used to index the material. The advantages for the library were obvious. First, twenty-five students ac- complished in a few weeks what might have taken one or two librarians months to complete. Second, the library was able to have the indexing done by people who were qualified in the field of reliability engineering. A spot check of the student work showed that it was completely com- petent. The use of subject matter ex- perts to help in the classification and cod- ing of reference material seems to be growing within many industrial and re- search organizations. Some companies ex- pect, or at least encourage, their technical people to offer assistance to the technical iarity with automated reference systems may prove to be a valuable if not neces- sary asset in the education of a graduate engineering student. The President's Science Advisory Committee strongly urges that engineering students be made to acquire some proficiency in literature techniques and information retrieval. 1 Perhaps the technical libraries in our en- gineering schools can help to meet this goal by devising projects similar to the one described above. If so, both the library and the student will have gained. • • 1 Science, Gove1·nmettt, and Information: Th e Re- sponsibilities of the Technical Community and the Government in the Transfer of Information. A Report of the P .resident's Science Advisory Committee (Wash- ington: Government Printing Office, 1963), p.28. Additional ACRL Appointments RoBERT C. MILLER remains as chairman of the Law and Political Science Subsection; Bruce M. Brown was reappointed to the Publications Com- mittee; Andrew J. Eaton has been appointed chairman of the St. Louis Conference Local Arrangements Committee with John B. Abbott and Carl H. Sachtleben appointed as members; Mrs : Margaret K. Toth continues as editor of the ACRL Microcard Series with the Editorial Board consisting of Felix Reichmann, newly appointed to managing editor, Thomas R. Buckman newly appointed, and E. Judson Humeston, Jr. • • 466 COLLEGE AND RESEARCH LIBRARIES