Rising periodicals costs have been an ongoing reality in academic libraries for more than a decade. In one of many early articles on the subject, Marcia Tuttle explained the complexity of the problem from the diverse points of view of librarians, publishers, subscription agents, and college and university faculty (Tuttle, 1989). Year by year, since the late 1980s, libraries have tried to deal with the problem by "tinkering" with their serials lists, sometimes without realizing the fact that a thorough rethinking of serials commitments might be necessary. Large universities whose mission is to establish and maintain research level collections began to economize by eliminating duplicate titles and implementing other cost-cutting measures. Much experimentation and some research has been done to test the notion of sharing expensive titles within a region or within a system through telefacsimile ("Telefacsimile...," 1988).

As difficult as the "serials problem" was then, it got worse as we entered the 1990s. By 1991, most academic librarians were well aware of the problem, and some Association of Research Libraries (ARL) members began to make concerted efforts to exert political pressure by large scale cancellations. Reports in the Chronicle of Higher Education on this subject raised the awareness of faculty members, when they read that cancellations averaging $140,000 by ARL libraries were not unusual (Nicklin, 1991).

Small to middle-sized academic libraries (those serving 500-6,000 students), which have as their mission supporting mostly undergraduate programs, have also experienced profound problems making their funding extend to cover their serials list (Frazer, 1992). In addition to this problem, we in smaller institutions have sometimes found it difficult to establish exactly which are the most important periodicals to have in our libraries to support our particular undergraduate level curricu-lum. Standard lists help, but each institution's mix of programs is slightly different. When one is dealing with a total possible list of under 2000 titles, picking the best 2000 for us is an important and challenging collection development effort. Nonetheless, when the resources do not extend far enough to cover the costs of maintaining subscriptions, college and university libraries of all sizes must look for ways to balance the budget while doing the least damage to the educational life of their institutions.

By 1992, increasing costs came crashing into dramatic focus for many small to midsize aca-demic libraries, when they met reduced materials allocations head-on. In public institutions and private ones, overall institutional budgets have been reduced because of the recession in the economy at large. Sometimes the cause of the reduced college budget is decreased revenue from eroded en-rollment; sometimes the cause is reduced allocations from legislatures.

After three years of midyear materials budget reductions at our institution (a small, compre-hensive, publicly-supported college), it became clear that we needed to make substantial cuts in our periodicals budget. We needed to reduce our serials expenses, since, by virtue of the reductions coming unexpectedly in midyear, our monographs allocations had been greatly affected by the cuts. Since money for serials had already been expended, the only area available to be cut was the mono-graphs budget. We knew that in order to retain resources to purchase monographs from our opera-ting budget in future years, we needed to make hard and careful choices immediately. We were forced to cut $60,000 from our serials budget in the spring and summer of 1992. As you may guess, this was a stressful and difficult time for us.

A careful and deliberate consultative process was begun with academic departments through our Collection Development Committee to make the best decisions possible regarding deletions. Meanwhile, other options were explored for locating the research-level materials that were needed by our 234 full time faculty and our 177 graduate students. The literature has many examples of articles about the process of performing a serials cancellation project, and although it is always a tense, difficult process, it is a well-documented one (Metz, 1992; Sapp & Watson, 1989; Segal, 1986). This paper, however, is intended to describe the strategic process used to introduce simple, unmediated online searching and "just in time" document delivery services to meet the information needs of faculty.

2. LOCATING ALTERNATIVES FOR RESEARCH-LEVEL JOURNALS

As the Collection Development Committee was pursuing its evaluation of our serials list,

keeping in mind the concept of refining the serials list so that it met the special needs of our particular undergraduate program, it was also necessary to begin the process of locating alternatives and planning how to fill the need for the research-level materials that we were eliminating. In conver-sations with the director of another library in the region, I was reminded of the burgeoning availa-bility of patron-searchable databases with associated commercial document delivery services. These now include OCLC's Article First, CARL UnCover 2, Faxon Finder and Faxon XPress, among others. While we had never before seriously considered such services, since we believed that they would be prohibitively expensive, the situation we were facing was serious enough for us to consider every possibility. We came to believe that simple, unmediated, online searching of the literature by faculty, followed by interlibrary loan and limited document delivery might meet our needs. Vendors of document supply services were evaluated, and CARL Systems of Denver, Colorado, was selected on a trial basis to provide our College's primary access to research-level periodicals, through UnCover 2. As stated earlier, we focused our own collecting efforts on those periodical titles needed by general undergraduate students.

3. STRATEGY FOR INTRODUCING THE NEW SERVICE

Since the purpose of this paper is to outline the strategy for introducing simple online searching and document delivery to the college community, it will take the administrative perspective. Consi-dering the political risks involved, and the potential for extremely negative relations with faculty during a serials review and cancellation project, it is important that the introduction of the new service be carefully planned and implemented.

It might be useful to look at how a library's services fit into a college's political environment during times of fiscal constraint. Obviously, each campus environment is unique, but some general observations may be shared. Our College Library is seen by some of our faculty much like a "utility." From this viewpoint one doesn't think too much about the Library, but, like the lights, water, and electricity, one expects that the service will be there, ready for use.

We're seen by some faculty as simply an arm of the College administration. This view places the Library in the same category with building maintenance, the office of admissions, or the coun-seling service. All of these are important campus areas, but their missions are not linked, as are the Library's mission and those of academic departments, to the curriculum.

Because of these two attitudes which sometimes exist among college faculty, cutbacks in the Library can serve to compound the effects on faculty of other expressions of fiscal constraint that are experienced in bad financial times. Faculty notice when the positions of colleagues who have recent-ly retired are not immediately posted for hiring. Vacant lines, especially those of incentives, often remain unfilled. The worst news yet for some departments is that the line may be reassigned to another department which now needs it more. Secretaries (another mark of status in academic departments) now serve two or three departments, where formerly each department could count on having its own secretary. Even paper and other supplies seem to be limited.

Now the Library's budget comes into the picture. The Library announces that it is cutting the very journals that one feels that one needs to do research. Some faculty, as we know, are very proprietary about "their" journals. The need to economize on journal titles can be the "straw that breaks the camel's back," with the Library seen as a "bad guy" on campus. When the Library has to cut titles to balance its budget, it is often one of many financial cuts that faculty are forced to endure. It is even possible that the Library may experience some scapegoating by faculty who feel more comfortable expressing their outrage at this state of affairs to the Director of the Library than to the Dean.

One of our goals, therefore, during the process of serials reductions and providing alternative document delivery services, was to stress the positive aspects of the new arrangement. The impor-tant feature, we became convinced after exploring the document delivery services offered, was that our graduate students and faculty could have better access to the periodical literature than had been possible with our earlier larger budget, let alone with the reductions we were facing. Even with the more generous budget, we would not have had such easy access to the more than 13,000 titles that were, at that time, available through CARL. It should be noted that in the one year from the intro-duction of our service until the present, CARL UnCover 2 has increased its access. Now, 14,000 periodical titles with more than four million individual articles are available, as is access to more foreign periodicals through an agreement with B. H. Blackwell, Ltd. as well as with the British Library Document Supply Center.

4. ADVANTAGES OF ONLINE ACCESS

The ease of searching on the CARL databases was another factor that gave us confidence that the online searching and document delivery option would be fruitful for us. Since CARL's search engine has been used since the late 1980s to search for books in the public library environment, it had been shown to be very useful for unmediated searches. CARL's search strategy was simple, yet powerful, so we were confident that faculty would easily be able to search on their own, from home, the office, or from the Library, and find the sources they needed.

Another CARL feature that we found especially compelling was the availability of journal tables of contents online, as each new issue is added to the remote library's holdings. Even without document delivery, this feature had great appeal to our faculty who had previously needed to make a sixty-mile trip to our nearest ARL library to find what was currently being published in journals peripheral to the researchers' primary reading lists.

5. GATHERING SUPPORT

Once we became convinced of the merit of the project, it was important to collect others' support. First, the Vice President for Academic Affairs was shown what we hoped would be the financial advantages of this solution to what he realized could be a faculty relations nightmare for him, as well as for the Library. He suggested immediately beginning demonstrations for Deans and Department Chairs.

This was easier said than done, but within a week we had obtained a working password and scheduled visits from Deans for demonstrations of the system. They were delighted! As their enthusiasm spread, we soon had Department Chairs and other influential faculty coming to the Library for demonstrations in their subject areas.

Persuading faculty leaders of the value of the project was an important next step. For this purpose, three "electronically savvy" faculty, one from the humanities, one from the social sciences, and one from the natural sciences, were approached individually in order to show them the system. Two of the three had personal computers with modems in their offices, so we visited them there to show them what would be available to them when our program was approved. Doing a demons-tration of new technology for an expert user often quickly turns from being a "show and tell" to being a "let me do it myself" exploration of the application. Amidst expressions of great enthusiasm, all three were delighted with the new source for information. Two of the three agreed to help us extend the awareness of the new service to others in their disciplines.

Back in-house, library faculty were making themselves acquainted with the new service, practicing searching, exploring the CARL databases, learning the power and the limitations of the system. Although not all of our eleven librarians took an active interest in developing searching skills, a cadre of practiced searchers developed, taking responsibility for being on call to assist faculty with their searches. Librarians developed searching guides and other handouts that described CARL and how to access the service using the terminals to our online catalog in the Library and personal computers from the office and home. Although not all librarians became active agents of change to promote the new system, a number of interested library faculty gave additional one-on-one demonstrations in the Library, and a few went to faculty departmental meetings to answer questions about the service. From an administrative standpoint, this outreach effort can be seen as an exten-sion of the work begun in the close liaison work of the serials review project several months earlier; a sip of sweet lemonade made from the sour bad news of financial crisis!

6. WORKSHOP FOR FACULTY

Sometimes faculty like to be shown a new technical skill by a colleague, rather than admit to someone outside the department that there is an area of weakness. To address the need for additional training among the faculty, and to offer it in a considerate way, we developed another idea.

Team teaching is fun, and can be very effective when done between library faculty and other faculty. We planned a workshop to be team-taught between one of our influential early-adopter faculty members and the Library Director. The workshop took place early in the academic year, and was a real success. That is not to say that there weren't the technical problems that are inevitable when you get thirty people in the room and attempt to have them do something with computers. But it was a success, because it exposed faculty to the new system and showed them that they could use it effectively with little training.

From an administrative point of view, the attendance was very interesting. Our college has 22 academic departments, and the workshop had garnered voluntary attendance from 16 of them, with representation about equal from the arts, humanities, and natural sciences. Attendance from the social sciences departments was about twice as high as that from any of the other areas, perhaps simply because more faculty are located in these departments. It should also be noted that the demonstration and workshop took place at 8 AM in a student computer lab (scheduled early to avoid displacing students), so the 26 persons registered had made a special effort to be there. Since there were so many interested faculty attending the workshop, and since their departmental representation ranged well across the curriculum, spreading the word about the new service became much easier. Faculty asked library staff for help, or they asked colleagues in their buildings.

Articles about the service appeared in the Library's newsletter, the official College weekly news publication, the local newspaper and even the student newspaper. At about this point, the service began to be expected and incorporated into that "utility" function of those library services which some of the faculty "take for granted." Although this is true, many of our faculty do feel lucky to have this service available, when they compare notes with colleagues at other institutions. Our librarians have also been invited to present formal and informal comments on our service regionally and within our State University of New York (SUNY) system.

7. USAGE LEVEL AND COSTS

Since July 1, 1992, we have spent approximately $7,100 on this project, of which approxi-mately $3,500 is directly attributable to CARL services. From July 1992 through June 1993, we requested 121 faxed articles from CARL at an average cost of $9.78. We paid $900 for each of two passwords, and then added another $100 to authorize our use of ERIC, when in November it became a restricted service. We established a deposit account from which our document delivery charges are deducted. In that account, approximately $2,000 remains. Besides these expenses, we have also included in our budget recharged interlibrary loan costs, some startup costs related to computer equipment and record keeping software, some DIALOG costs, and commercial parcel delivery services. We have budget figures on the amount of money spent for purchasing document delivery, but we have no way of knowing how many times faculty have used the Library's password for searching, and then requested the article through interlibrary loan. We can tell you, however, that we have requested 3,366 articles through interlibrary loan this year, an increase of 38.5% over figures for last year.

Other expenses were minimal: the cost of adding another telephone line at the reference desk to use to dial direct in case of network problems, the cost of buying UnCover 2 searching guides and producing local handouts. Contrasting our cost per article from CARL with others in the literature (Chappell, 1993), it seems clear that for us the option of using CARL was an excellent fiscal strategy.

Again, it is important to reiterate that we have purchased only two passwords and have not used the document delivery service for each citation retrieved. We've approached the process conservatively, choosing not to purchase a gateway, and to monitor the reported occurrence of password contention. This, fortunately, has not been a problem.

8. PROBLEMS

That is not to say that we have not experienced some problems in implementing our new ser-vice. We needed to work closely with Computing Services to configure our Library terminals to allow Telnet searching from the Library. This was not an overwhelming problem, but because of the short time-frame between our decision to implement CARL service and the return of the faculty to campus in the fall, we were under some pressure to make the technological changes as quickly as possible. Fortunately, Computing Services saw other advantages for themselves in making Telnet service available, so had already begun plans for their own implementation of that capability.

We continue to have some problems "getting the word out" about our service. Faculty some-times forget how to connect and search if they do not use the service regularly, so the need for retraining is not unusual.

User-friendly handouts were not all that easy to produce. Until the librarians became com-pletely familiar with CARL's search strategy and the services available, they were not comfortable in creating the necessary handouts. A corollary to that problem is that in this world, everything changes. It didn't take long before the handouts which we had prepared needed changes because of developments in Computing Services or at CARL. Sometimes, in fact, using the existing handout would cause the keyboard to lock, frustrating librarian and patron alike.

9. THE FUTURE

Now that we have a year of successful simple online searching and occasional document delivery, we are considering where to go from here. We believe that we may strengthen relation-ships with certain research libraries to facilitate interlibrary loan transactions for particular subjects. We may use other commercial online search and document delivery services to supplement some subject areas which librarians feel are weaker in CARL. And finally, we expect to do more publici-zing of our CARL services to students. Although we have had student users, we have not yet experienced a surge of student search requests.

10. THE GOOD NEWS

To end on a happy note, here follow a few comments from faculty, an administrator, and a student. The names have been omitted because of the potential political implications of one of the comments!
"This system, in my view, is more than simply a mechanism for identifying and obtaining relevant journal articles. It is also an outstanding research tool. The ability to do key word searches and to rapidly scan the tables of contents of the latest journals in my field would be incredi-bly useful to me even if it were not possible to actually obtain the articles by fax. The fact that articles can be obtained the next day via facsimile transmission eliminates the need to have many of these journals in our Library."

- from a faculty member

"I am very enthusiastic about this service. It provides, at minor cost, a service that im-proves scholarly access to material in a quantum-leap way -- and one that I, frankly, did not expect to see even considered in these grim times. In fact, in terms of increasing pressures on library finances, this . . . alleviates the grimness, if not giving one the opportunity for genuine joy, about what we may term 'scholarly access' to literature." - from a faculty member

"I find UnCover to be much more efficient than Biological Abstracts, which I had used many times prior to the installation of CARL. In addition to the efficiency of UnCover, it is also much more convenient than the abstracts. I am able to access UnCover from my department's computer lab, as well as from home via modem. This has given me a greater flexibility in the times and places I can work. Through UnCover, I am able to obtain all of the information I need to request papers from journals that are not available in Reed Library, and I have them within a week, rather than four to six weeks. In general, access to CARL has greatly improved my ability to complete my research for both my classwork and for my thesis. I am now able to work more quickly and efficiently."

- from a graduate student

"I strongly support the implementation of CARL. The expansion of access to source material as well as the speed and ease of access provides an outstanding educational tool for our campus and allows departments to prioritize and condense journal subscriptions. Given accelerating journal costs, the decision to implement CARL at SUNY College at Fredonia was one of the most cost-effective decisions I've experienced as an administrator."

- from an administrator

Making lemonade from lemons isn't necessarily difficult. If you're presented a sour situation, first add a bit of ice to cool things down a little. Then look for an alternative that will quench the faculty's thirst for bibliographic information. Sweeten the mix with some good-hearted optimism by service-oriented librarians, blend carefully with a few well-spent dollars from the Administration, and voilá! A refreshing transformation!

ACKNOWLEDGEMENT

The serials review project at Daniel A. Reed Lirary, and our subsequent online searching and document delivery service were the result of the concerted efforts and positive contributions of many at the Library and the College. Special thanks go to Richard Hart, the Chair of our Collection Development Committee, Margaret E. Pabst, our Interlibrary Loan Librarian, all of the fine librarians at Reed Library, and the College's Computing Services personnel. As well, I am indebted to George Telatnik, of Canisius College of Buffalo, New York, for his good counsel as we explored document delivery services.

REFERENCES

Chappell, M.A., "Meeting undergraduate literature needs with ILL/document delivery," Serials Review, 19 (1): 81-94 (Spring 1993).

Frazer, S.L., "Impact of periodical cost escalation on small and medium-sized academic libraries: A survey," Journal of Academic Librarianship, 18 (3): 160 (July 1992).

Metz, P., "Thirteen steps to avoiding bad luck in a serials cancellation project," Journal of Academic Librarianship, 18 (2): 6-82 (May 1992).

Nicklin, J.L., "Libraries drop thousands of journals as budgets shrink and prices rise," Chronicle of Higher Education, 38 (16): A29 (December 11, 1991).

Sapp, G. and P.G. Watson, "Library-faculty relations during a period of journals cancellations," Journal of Academic Librarianship, 15 (5): 285-289 (November 1989).

Segal, J., "Journal deselection: A literature review and application," Science & Technology Libraries, 6 (3): 25-42, (Spring 1986).

"Telefacsimile and scanning technologies," unpublished final report of H.E.A. Title II-D grant recipients: SUNY, University Centers Libraries (1988). Available from Stephen Roberts, SUNY, University at Buffalo, 432 Capen Hall, Buffalo, NY 14260.

Tuttle, M., "The high cost of scholarly journals: Working together to find solutions," Library Issues, 9 (4): 1-2, 4 (March 1989).