College and Research Libraries " Humanists and Electronic Information Services: Acceptance and Resistance Stephen Lehmann and Patricia Renfro The design and implementation of online systems in libraries have proceeded without much demonstrated empirical understanding of the complexities of users' needs. The authors interviewed a group of humanist scholars at the University of Pennsylvania about their experiences with the RUN database. These interviews yielded a wide range of responses. Four factors emerged consistently as significant determinants of use: content, connectivity, user- friendliness, and cost. The significance of these factors, individually and in relation to one another, is discussed and evaluated. n f the development of online systems gives librarians a new opportunity to reshape - library service, it also man- dates a renewed commitment to under- standing how scholars work and how they use information. Librarians under- stand the use of online catalogs, journal article databases, and now full text re- trieval systems only intuitively, yet they are puzzled if scholars do not fall upon these resources with enthusiasm.1 What will draw researchers to compu- terized information systems, and what will repel them? What are the trade-offs of the new technology, and how do these stack up? Where should resources be al- located in the face of difficult choices? At the University of Pennsylvania, we recently had an opportunity to ask these questions. After making personal search accounts to the RLIN database available to faculty at no charge, we interviewed a small group of humanists who had be- come committed RLIN searchers.2 Know- ing that this catalog of the holdings of major research libraries could be an im- portant resource for any scholar in- volved in historical research, we wanted to know why it worked for some, but not for others. What distinguished enthusiastic RLIN searchers from colleagues who had little interest in opening search accounts or who, having received accounts, rarely used them? Was the database inade- quate for some purposes? What were the barriers to its use? This is a case study in the use (and nonuse) of new electronic resources by humanist scholars. The numbers are small and the evidence is anecdotal, but what we heard was suggestive and even compelling. As we talked with this group of scholars, certain themes began to recur. We were reminded of the fun- damental importance of database con- tent. We became a ware of a range of issues relating to connectivity. We heard comments on the importance of user- friendliness and reactions to the issue of cost. We believe that the insights these interviews gave us provide useful point- Stephen Lehmann is Humanities Bibliographer and Patricia Renfro is Associate Director of Libraries, Public Services, at the University of Pennsylvania, Philadelphia, Pennsylvania 19104. The authors gratefully acknowledge the help and cooperation of the faculty members of the University of Pennsylvania's School of Arts and Sciences who participated in this project. 409 410 College & Research Libraries ers for the planners of academic informa- tion environments and hypotheses about the use of computers in the humanities. These hypotheses can later be tested and reformulated as the new generation of local mega-systems starts to provide data about their use. CONTENT Librarian: His RLIN central to your work?" Philosopher (and heavy RLIN user): ~'Thinking is central to my work." The most fundamental distinction be- tween researchers and librarians is per- haps the emphasis on content by the one and on access by the other. Generally, the concerns of librarians-information or- ganization, control, and access-hold the same kind of interest for scholars as a car does for family vacationers: it's what gets you there. This separation of process (technique, technology) from con- tent accounts largely for librarians' failure to excite teaching faculty about library in- struction and also explains, at least in part, their lack of interest in online searching. Librarians marvel at the retrieval power of online search systems, Boolean capa- bilities, keyword searching, and the rest of it, but the humanist scholar, after checking for his or her own publications, looks for that seminal work published in Belgium in 1937 and wonders what the use is of a system that does not in- clude it. Although RUN's coverage of nine- teenth- and early twentieth-century im- prints is far from complete, the scholars we interviewed clearly felt there was enough in the database for it to be useful. RUN statistics show that about 49 per- cent of the titles in the books and serials files were published after 1970 and that 8.4 percent of the titles were published in the nineteenth century. 3 We conducted a small study using a bibliography of Hegel scholarship consisting largely of titles not in English and found that 37 percent of the titles published from 1844-1879 were in RUN, as were 49 per- cent of those published from 1880-1912, 53 percent for the period 1913-1945, and 77 percent for 1946-1975. Inclusion of September 1991 titles from the largely English-language listings of Annals of American Literature was 84 percent for 1817-1843 imprints and 91 percent for 1844-1879.4 Files that go back twenty or even forty years are useful, but the work of scholars is con- tent-driven, and in the humanities con- tent knows no barriers of time or language. Without exception or hesitation, every fac- ulty member we interviewed expressed a strong preference for retrospective expan- sion of the RUN database over further development of a research-in-progress file. The most fundamental distinction between researchers and librarians is perhaps the emphasis on content by the one and on access by the other. For at least one Penn faculty member, the relative depth of the RLIN database makes it valuable not only for research, but also for teaching. His students search RUN on his office PC to find what has been published on a given topic. The "extraordinary riches of the database," he maintains, give students a sense of wonder: "RUN is a teaching tool in a deep sense ... through RLIN [students] realize that they are becoming a part of a transcultural, transtemporal commu- nity of inquiry." Unlike their professors, students tend to take RLIN's power for granted; what impresses them, rather, is the content of the database, much as the printed catalogs of the great national li- braries impressed earlier generations of scholars. CONNECTIVITY Penn faculty member: HI want the infor- mation right away, before I write the next paragraph." The idea that information should be accessible at the scholar's workstation, whether in the home or office, is funda- mental to the concept of the electronic scholar. While RLIN has proven its use- fulness at the reference desk (logging approximately 40,000 searches a year at Penn), the direct search accounts have allowed scholars to use RLIN in a differ- ent and extremely powerful way. With- out the barriers imposed by the need to be in the library and to ask a librarian to mediate a search, use of RLIN has changed. The autonomous relationship between the searcher and the database encourages relaxed browsing-intellec- tual cruising, one searcher called it. A member of the Penn English Department spoke of the "gigantic difference" that direct access makes because it allows for "the browser-shopper frame of mind that comes from years of being social- ized in the library stacks." RLIN satisfies that urge and provides "a sense of seren- dipitous exploration." As one scholar ex- plained, "I won't share my semi-focused curiosity with a librarian." The electronic information resource must, then, be available wherever re- search or teaching goes on. Unfortu- nately, this is not yet always the case. Some people can write only at home; others find access in campus offices to be vital; others need to be away from both home and office, perhaps in a library study carrel. Scholars are limited by the hardware and software available to them. Humanists typically have fewer resources than scientists to pay for cam- pus network connections and for the hardware and software that will maxi- mize their use of the network. Is institu- tional support for home modems and PCs adequate? If the network and online catalogs are designed to operate virtu- ally twenty-four hours a day, subsidized computers and modems might make the institutional investment in a twenty- four-hour resource worthwhile. But connectivity means more than the installation of network connections and the provision of the necessary hardware and software. It means connecting peo- ple to resources-in other words, getting the scholar to the resource with a mini- mum of effort on his or her part. Many scholars can set up and configure hard- ware and software, find · their way through systems manuals, and persist in eliminating bugs. These are. the com- puter literate members of our faculties, people who jumped at the opportunities Electronic Information Services 411 offered by new technology and had the interest and inclination to try it out. All of Penn's frequent RLIN searchers are in this category. Most scholars, however, do not have the time, inclination, or en- durance for this process. One searcher talked about the "endemic impatience" that humanists have for computing. Many described the fear and frustration that they see in their colleagues. As information pro- viders, librarians must deal with the entire range of reactions to technology, from the enthusiasm of the humanist hackers, to the hostility of "constitutional Luddites," to the impatience of the average-to-busy, overstressed researcher whose time is al- ready budgeted tightly. The major· problem identified by all the libraries that participated in theRe- search Libraries Group's (RLG) 1988 Re- search Access Project to experiment with direct scholars' use of RLIN was the issue of connectivity-people with ade- quate equipment, but no support for in- stallation and setup.5 The same picture emerged on all campuses: faculty inter- ested in trying out the database, but frus- trated by the difficulties of adapting hardware and software to access it. In order to realize their investment in net- works, universities must find ways to pro- vide scholars with simple, universal gateways. Some libraries are beginning to meet this need by offering their pa- trons a straightforward menu of infor- mation options-an online catalog, locally loaded databases, and general li- brary information. USER-FRIENDLINESS Penn faculty member: "RLIN is difficult to manipulate . ... It has the appearance and reputation of complexity." Another Penn faculty member: "I would rather put the money into making the system more powerful than into making it more user-friendly." Although the RLIN search interface is logical and consistent, it is not remark- able for its ease of use. The faculty to whom we spoke, themselves RLIN users and also motivated, patient, and gener- ally confident computer users, agreed that many of their colleagues, perhaps 412 College & Research Libraries especially humanists, would find the less-than-intuitive commands and codes an impediment. Certainly, we encoun- tered faculty who seemed frustrated, even at first exposure to the command structure. The major problem identified by all the libraries ... was the issue of con- nectivity-people with adequate equipment, but no support for instal- lation and set up. In the absence of hard data, we cannot say with any certainty how significant a factor user-friendliness is relative to con- nectivity and cost, but we surmise that where all other circumstances are favor- able, an interface like RLIN' s should not be a deterrent in most cases. Where other disincentives, such as poor connectivity, interfere, an unintuitive interface will be a ready-at-hand reason not to go further. It is also evident that a relatively un- friendly interface discourages occa- sional users-scholars not engaged in ongoing, intensive research requiring a large bibliographic database, but rather faculty who need now and then to go beyond the local catalog to verify a title or determine a location. COST Librarian:~'Would you use RUN if you had to pay for it?" Historian: (with raised eyebrow) #How much?" Librarians know from experience that library services are cost sensitive. In some disciplines researchers can pass along costs and charge them to grants, but this is generally not the case in the humanities. The committed RUN searchers at Penn indicated that the database had become so important to their work that they prob- ably would be prepared to pay something toward the search costs. They probably would not have allocated funds initially, however, before acquiring some knowl- edge of the system's usefulness. Clearly, direct charges are a significant barrier to use in some academic environments-at September 1991 most institutions, travel costs, photocop- ying fees, and research assistant salaries all compete for the limited discretionary research dollars available to human- ists-and because cost is often related to the volume of use of a system, the pro- cess can be self-defeating. CONCLUSION This ranking--content, connectivity, user-friendliness, and cost-is, for the most part, relative and not absolute. That is, if a particular system happens to be extraordinarily user-friendly, poor con- nectivity might not significantly impede its use. Content, we believe, is a sine qua non, however accessible, easy to use, and cheap a system may be otherwise. How- ever, content, though necessary, is not sufficient. RUN searchers at Penn are faculty whose work benefits from the database but who also have enough comfort with computers to overcome hard ware and software obstacles. We are convinced that if more faculty had easier access to RLIN and that if the RUN in- terface were more intuitive or offered a menu-driven option, use of RUN would be much more widespread. If, on the other hand, libraries were to pass back costs, use would be significantly less. Every faculty member we inter- viewed expressed a strong preference for retrospective expansion of the RLIN database over further develop- ment of a research-in-progress file. While librarians will, in time, learn much more about the scholarly use of computers and computerized information systems than we have outlined here, this effort to understand will always be like shooting at the proverbial moving target. Surely electronic systems will engage scholars and librarians in a dynamic of change, where both sides interact in a back and forth of stimulation and adaptation. New technologies will spur on new re- search methodologies, and these, in turn, will guide new technological develop- ments. It is important that technology in Electronic Information Services 413 the scholarly environment be under- stood as a part of this dynamic process and not as a Darwinian, adapt-or-die im- perative. For this model to work, librari- ans cannot let systems be systems driven. Rather, decisions always should be in- formed by users' needs-in all their complexity. REFERENCES AND NOTES 1. See, however, Christine L. Borgman, Donald 0. Case, and Dorothy Ingebretsen, "Uni- versity Faculty Use of Computerized Databases: An Assessment of Needs andRe- sources," Online Review 9:307-23 (Mar. 1985); Donna Cornick, "Being an End-User Is Not for Everyone," Online 13:49-54 (Mar. 1989); Jan Homer and David Thirlwall, -.,.._ "Online Searching and the University Researcher," Journal of Academic Librarianship 14:225-30 (Sept. 1988); and Peter Stem, "Online in the Humanities: Problems and Possibilities," Journal of Academic Librarianship 14:61-64 Ouly 1988). 2. See also Fred Muratori, "RUN Special Databases: Serving the Humanist," Database 12:48-57 (Oct. 1990) and "Introducing Personal Searching to RUN," Research Library Group News 22:11-13 (Spring 1990); and Stephen Lehmann and Patricia Renfro, "Hu- manists at the Keyboard: Faculty Use of the RUN Database," Computers and the Humanities (forthcoming). 3. Research Libraries Group, Operations Update, 54:4 Oune 1990). 4. Lehmann and Renfro, "Humanists at the Keyboard." 5. For a description of the Research Access Project, see Melanie Dodson, "Faculty Access to RLIN at New York University: RLG's Research Access Project," C&RL News 49:522-23 (Sept. 1988). your collection measure up? You need reliable quantitative data to justify collection management decisions. You need a flexible analysis system designed with your library's goals in mind. Introducing ........................ OCLC/ AM/COS Collection Analysis Systems. Collection Analysis CD compares your holdings against those of similar institutions, using a subset of the OCLC database on compact disc. Tape Analysis gives you a custom-designed MARC tape analysis for your library or group. A Tape Match against Books for College Libraries is also ·offered. · . '' OCLC/ AMIGOS Collection Analysis Systems Available exclusively in the U.S. from AMIGOS Bibllographic Council, Inc. 11300 North Central. Expressway, Suite 321 Dallas, Texas 75243 (800)843-84~2 (214)750-6130