herring.p65 Journal Literature on Digital Libraries 39 Journal Literature on Digital Libraries: Publishing and Indexing Patterns, 1992–1997 Susan Davis Herring This study examines publishing and indexing patterns in the area of digital libraries to identify the core journals and the most effective de­ scriptor terms for online searching. A group of 232 citations on digital libraries from 1992 to 1998 was analyzed. Results show that none of the 105 journals listed has emerged as the leading title in the field and that, although two-thirds of the citations were identified by one of two phrases, indexing is inconsistent and subject searching unreliable for full retrieval. The study confirms that skill, persistence, and wide reading are required to keep up with developments in this evolving field. he number of journal articles published on the topic of digi­ tal libraries has proliferated in the past several years. Unfor­ tunately for those attempting to keep up with the subject, the literature is scattered across a number of disciplines, primarily library and information science and com­ puter science. Although it is difficult for specialists in any one field to be familiar with the literature in other disciplines, such shared knowledge is necessary for the successful development of new infor­ mation technology. Using the same ter­ minology and understanding the same mental models are necessary if computer scientists, information scientists/librar­ ians, and subject specialists are to coop­ erate in the advancement of digital librar­ ies. One of the best ways to build this understanding and common knowledge is through joint familiarity with the ma­ jor journals that contribute to the litera­ ture in the digital libraries field. The current research study was de­ signed to explore publishing and index­ ing patterns in the area of digital librar­ ies. The major purpose of the study was to identify the core journals, those that publish the majority of articles on the topic, in order to help professionals fo­ cus their time and attention on the most prolific journals. A second purpose was to examine the descriptor terms used by the organizations that index the articles on digital libraries, to both identify the major terms and reveal inconsistencies in indexing. The study was limited to sub­ stantive journal articles, including empiri­ cal, theoretical, and practical studies and excluding book reviews, product reviews, and directory lists. Literature Review A review of the literature revealed that no study of publishing patterns in digital library literature similar to the current study has been published to date. How- Susan Davis Herring is an Associate Professor of Bibliography and an Engineering Reference Librarian in the M. Louis Salmon Library at the University of Alabama in Huntsville; e-mail: herrings@email.uah.edu. 39 mailto:herrings@email.uah.edu 40 College & Research Libraries January 2000 ever, the methodology used here was de­ tailed by F. W. Lancaster and Ja-Lih Lee in their discussion of the use of online databases to track emerging issues.1 They demonstrated the usefulness of conduct­ ing online searches in groups of subject- specific databases as a means of tracking the growth of journal literature on par­ ticular topics and identifying the core journals in the field. A similar methodol­ ogy was used by Terrence A. Brooks, who conducted an analysis of the literature in superconductivity through fifteen years of INSPEC files to identify the core jour­ nals in that field.2 Chun Wei Choo noted that the success­ ful development of information systems requires coordination among domain ex­ perts, information technology experts, and information experts; this could be applied to digital library development as well.3 In a 1997 paper, Lisa Dallape Matson and David J. Bonski stressed the importance of communication between library and computer specialists working on digital library development.4 They also pointed out the dichotomy between the vocabularies used in library science and computer science literatures, which can easily lead to misunderstanding and a breakdown in cooperation. An earlier citation analysis by this re­ searcher demonstrated the lack of cross- fertilization in the literature on the de­ velopment of Internet search engines.5 Over 40 percent of the articles analyzed appeared in computer science journals, whereas only 20 percent of the articles studied appeared in the traditional li­ brary literature. Over 80 percent of the computer scientists writing on the sub­ ject published only in computer science journals. Patrick Wilson has documented the common practice of ignoring relevant literature when it does not fall within the defined discipline.6 However, in the bur­ geoning and highly interdisciplinary field of electronic information access, limiting oneself to the literature of one’s own field could lead to the neglect of a great deal of applicable and appropriate literature. The Study The current study focuses on a body of citations on the subject of digital libraries that were identified through computer­ ized searches in a variety of electronic indexes. The study was limited to the period from 1992 through 1997. Earlier research in the field indicated that, prior to 1992, little had been written other than theoretical essays. Because the data were gathered very early in 1999, 1997 was the last year in which complete indexing might be expected. Two research questions were posed for this study: 1. What are the leading journals in the area of digital libraries? For the purpose of this research, “leading journals” are defined as those that publish the highest number of articles on the topic. 2. Which indexing terms return the largest number of and most relevant ar­ ticles on digital libraries in electronic in­ dexes? Data were gathered using a selected group of electronic indexes to get a broad and representative selection of the litera­ ture. Searches were run in Library Litera­ ture, Humanities Abstracts, Social Sciences Abstracts, Wilson Business Abstracts, Gen­ eral Science Abstracts, Applied Science and Technology Abstracts, Readers’ Guide Ab­ stracts, and PAIS International. It is recog­ nized that results from searches in these databases are not comprehensive; how­ ever, they will identify articles from the major professional and popular journals that are most likely to be available to read­ ers and researchers in the United States. The search words “digital” or “virtual” or “electronic” adjacent to “library” or “libraries” were used in order to cover all possible terminology. Book reviews, product reviews, and directory listings of Web sites were deleted, and duplicate records were combined, leaving a total of 232 articles. The data were entered into an Excel spreadsheet for ease of analysis. Data Analysis The initial question concerned which journals were the major publishers of ar­ Journal Literature on Digital Libraries 41 formation Technology and Libraries be­ comes the leader, with ten articles in the six-year period. The steady in­ crease in the number of articles on digi­ tal libraries being published is obvi­ ous from the data, as shown in table 2. The higher number for 1995 reflects the special issue of Communications of the ACM mentioned above. An additional twelve journals each published three or four articles in­ dexed during this period, including Journal of the American Society for Infor­ mation Science (with four articles), D- Lib, College and Research Libraries, Infor­ mation Processing and Management, Li­ brary Trends, Online, Program, and Sci­ entific American (each with three ar­ ticles). Some of the journals coming in with two articles are ACM Transactions on Information Systems, Computers in Libraries, Electronic Library, Library Hi Tech, IBM Systems Journal, and Searcher, as well as Forbes, Fortune, and Science News. Inconsistencies in indexing became obvious as the data were analyzed. A summary of the most frequently used subject headings/descriptors appears in table 3. Despite the fact that all but one of the databases searched are pro­ duced by the same provider, indexing terms varied remarkably across the databases. Even the primary descrip­ tor for digital libraries differed. Library Literature uses “Virtual Library,” Wil­ son Business Abstracts and Applied Sci­ ence and Technology Abstracts use both “Digital Libraries” and “Virtual Li­ brary,” and Readers’ Guide Abstracts, General Science Abstracts, and Social ticles on digital libraries. Of 105 publica- Sciences Abstracts use “Virtual Libraries,” tions listed, only forty-five (42.8%) pub- as does PAIS. Of the 232 articles analyzed, lished more than one article on the topic 158 were identified with one or more of during the six-year period studied. The these three terms; 74 were indexed using top eleven, each of which published five other terms entirely. or more articles during that period, are The number of terms assigned to ar­ listed in table 1. Twenty-two of the ticles varied from a low of one to a high twenty-four articles published by the top- of eight, with a mean of 2.82 and a mode ranking journal, Communications of the of two. Thirty articles received only one ACM, appeared in one special issue in term; the majority of these were either 1995. Discounting that special issue, In- “virtual library” or “digital libraries.” 42 College & Research Libraries January 2000 TABLE 2 Articles Published per Year, 1992-1997 1992 1993 1994 1995 1996 1997 total 10 20 19 62 56 65 232 edgeable about cur­ rent issues would be wise to scan Informa­ tion Technology and Libraries, Database, Computer, Science, Other single terms included “libraries,” “Internet,” and, perhaps most surpris­ ingly, “intellectual capital.” Discussion The data clearly show the increasing rate of publication of articles on digital librar­ ies. More than six times as many articles were published in 1997 than were pub­ lished in 1992. The simple proliferation of articles makes it more difficult for the professional to maintain awareness of re­ search in the field. Moreover, this diffi­ culty is compounded by the broad range of publications in which the literature appears. Analysis of the data gives no clear an­ swer to the first research question, What are the leading journals in the area of digi­ tal libraries? It is obvious that no single journal can be considered the leader in publishing these articles. At the least, the professional wishing to remain knowl­ and the two stan­ dard library science journals, American Libraries and Library Journal, on a regular basis. Communications of the ACM, an impor­ tant scholarly journal in computer sci­ ence, only occasionally includes articles dealing with the topic. However, when they appear, the articles are important, numerous, and in-depth, and well worth reading. Science is another well-respected and scholarly journal that is worth regu­ lar perusal by the librarian, information specialist, or computer scientist. The Jour­ nal of Academic Librarianship is the only li­ brary science title on this list that special­ izes in research articles. The Journal of In­ formation, Law, and Technology also is po­ tentially useful because of its legal focus. The alert reader will immediately no­ tice that this list does not include some journals that might be expected to appear. The outstanding example is D-Lib, the electronic journal that focuses specifically on advances in digital libraries. However, TABLE 3 Indexing Terms and Occurrence Rate Index Term 1992 1993 1994 1995 1996 1997 total Subject-specific Terms Virtual Library Virtual Libraries Digital Libraries 1 1 8 1 4 8 2 3 15 3 27 27 4 10 24 5 23 83 15 68 Other Terms Libraries 1 Information Systems 1 Internet Electronic Publishing Automation of Library Processes 2 Information Retrieval 2 World Wide Web 2 1 4 1 2 1 2 5 5 8 4 2 5 2 6 6 5 8 5 1 4 15 11 9 6 2 5 28 27 24 18 13 12 11 Journal Literature on Digital Libraries 43 of the databases searched, this journal was indexed only by PAIS International. PAIS began selective indexing of D-Lib in 1997, although the journal started publication in 1995. Only three articles from D-Lib were identified in this study. The simple difference between “library” and “libraries” could make the difference between an adequate search and one that is barely success­ ful. The second research question con­ cerned the most effective indexing terms. The results of this analysis indicate that two phrases, at a minimum, are required for reasonable results. Using only “Digi­ tal Libraries” would have returned less than one-third of the total; and “Virtual Libraries” alone would have returned less than 10 percent. Even broadening the search and truncating the descriptor terms to “Virtual Librar* OR Digital Librar*” would miss almost one-third of the final results. This inconsistency in indexing ter­ minology could present a serious problem for the naive searcher who is unaware of the various possibilities or the specialist who thinks in strictly defined terms. Not all users understand either boolean opera­ tors or truncation. The simple difference between “library” and “libraries” could make the difference between an adequate search and one that is barely successful. Perhaps more surprising was the num­ ber of articles that received only one in­ dexing term. Although these comprised only 13 percent of the total, it is still dis­ appointing that the indexers found no other, perhaps more useful, terms. Conclusions The digital library is a new phenomenon, and the present study has confirmed that the literature on this topic is still evolv­ ing. As is normal in other advancing fields, indexing terminology is still evolv­ ing and common terms are only begin­ ning to be established. Clearly, the per­ son wishing to remain current on the subject must cast a wide net in an attempt to retrieve relevant publications. The study described here focused on only one part of the literature in this field. A great deal of the discussion and re­ search on digital libraries is appearing outside the traditional journal format, primarily in conferences and online pub­ lications. It might be illuminating to con­ duct a study similar to this one examin­ ing the publication patterns and means of access to conference proceedings and online papers. Notes 1. F. W. Lancaster and Ja-Lih Lee, “Bibliometric Techniques Applied to Issues Management: A Case Study,” Journal of the American Society for Information Science 36, no. 6 (Nov. 1985): 389–97. 2. Terrence A. Brooks, “Core Journals of the Rapidly Changing Research Front of ‘Supercon­ ductivity,’” Communication Research 16, no. 5 (Oct. 1989): 682–94. 3. Chun Wei Choo, Information Management for the Intelligent Organization: The Art of Scan­ ning the Environment. (Medford, N.J.: Information Today, 1995). 4. Lisa Dallape Matson and David J. Bonski, “Do Digital Libraries Need Librarians?” Online 21, no. 6 (Nov.–Dec. 1997): 87–92. 5. Susan Davis Herring, “The Value of Interdisciplinarity: A Study Based on the Design of Internet Search Engines,” Journal of the American Society for Information Science 50, no. 4 (Apr. 1999): 358–65. 6. Patrick Wilson, “Unused Relevant Information in Research and Development,” Journal of the American Society for Information Science 46, no. 1 (Jan. 1995): 45–51.