Jay Jordan, OCLC's president and CEO, opened the symposium with a brief overview of the history of collaboration between LC and OCLC, particularly in the realm of building WorldCat into a 45-million-record resource. Following Jordan was Winston Tabb, LC's associate librarian for library services, who highlighted LC's and OCLC's milestone anniversaries. Next, Diane Nester Kresh, LC's director of public service collections, introduced the program and the speakers.
 

A Hot Topic
R. David Lankes, director of the virtual reference desk and assistant professor at Syracuse University's School of Information Studies, reported on the current status and future directions of digital reference. He cited a recent study by Joseph Janes that found that 97.3 percent of academic libraries have Web sites, and 44.7 percent have digital reference services—defined as a mechanism by which people can submit their questions and have them answered through e-mail, chat, or Web form.

Lankes noted that virtual or digital reference is becoming a hot topic for conferences. Now is the time to institutionalize digital reference by creating common technical and quality standards and by developing software built for both real-time and asynchronous reference. Lankes and Charles McClure, a professor at Florida State University's School of Information Studies, have already collaborated with OCLC in one research project to identify standards (http://www.quartz.syr.edu/quality). Urginglibrarians to make their expertise part of their collection, Lankes said that commercial search services, such as AskJeeves, are not the enemy and won't displace librarians as long as they are committed to quality. The "Librarian Ascendant"—the librarian of the future—must come to terms with his or her new roles as filters and information consultants.
 

Pasadena's MCLS
Representing the Metropolitan Cooperative Library System (MCLS) in Pasadena, California, reference coordinator Susan McGlamery provided an energetic account of and a few war stories about cooperative reference using e-commerce software. Awarded a grant in 1999 to purchase software, this multi-type consortium of more than 40 libraries aimed to provide 24/7 service, live interaction with a librarian, a network of subject specialists, and point-of-need Internet service. It selected Web contact center software that provides live collaboration, routing, reporting, request tracking, scripting, authentication, and browser compatibility.

McGlamery displayed a number of system screens via the Santa Monica Public Library's Web site (http://www.smpl.org/library) to show how the digital reference system actually looks. The consortium is now concentrating on Web-accessible information, pushing Web pages to users, and building FAQs. During the process of developing digital reference processes, librarians identified a number of problems. One issue was the increasing level of computer skill and technical expertise required of them. McGlamery concluded by stating that chat reference is just in its infancy and has a long way to go before being as easy and spontaneous as an in-person reference interview.
 

Santa Monica Public Library
Nancy O'Neill, principal librarian at the Santa Monica Public Library, followed up with her session, "How Digital Reference Works in the Public Library." Despite her library's vision of using digital reference services to meet the needs of a diverse clientele wherever they are, O'Neill stated that 24/7 service is not quite there yet. She suggested that before initiating such a venture, librarians need to consider how digital reference service fits into the library mission and institutional culture, whether there is acceptance from administration and staff, and whether the appropriate technology infrastructure exists.

Having provided an e-mail reference service since 1989, in March 2000 Santa Monica Public Library began participating in the Collaborative Digital Reference Service (CDRS) through the Library of Congress initiative. In July it started interactive reference through the MCLS's 24/7 Reference Service. O'Neill described the professional librarians who participated as fearless risk-takers. The library learned lessons that included rethinking the reference interview process, frequently updating written policies, coordinating digital reference with other reference services, evaluating for quality control, and developing new competencies. O'Neill defined digital reference as a core service, not an add-on, and said, "When our library clients are out there on the Web, I want to be out there to meet them."
 

Cornell's Libraries
Cornell University's Paul J. Constantine, head of reference services at the Olin Kroch Uris Libraries, followed with an academic case study. Beginning with the caveat that his library system represented only one way of handling digital reference that might not work for others, he explained that the system is constantly experimenting and refining and doesn't have all the answers. With digital reference there's a lot to learn about user behavior, user needs, and cost models.

Physically, the Olin Kroch Uris Libraries comprise three libraries, but they are virtually connected with one reference services division. Constantine said that Cornell had developed a plan for 2000­2002 with the goal of ranking in the top 10 of Association of Research Libraries (ARL) members. The plan's vision statement includes the ideas that users must have 24/7 access to high-quality digital reference services from anywhere, reference service must be a key component of the digital library, and digital reference should be "high-tech and high-touch." In-person reference transactions have dropped significantly at Cornell and at other research libraries. Constantine said that we have to move to digital reference because access to our collections has no time or space constraints, users of these resources continue to need assistance, and use of competing services is escalating. (For example, there are more than 4 million queries per day on AskJeeves alone.)

During the '90s, Cornell initiated e-mail reference and online tutorials, experimented with videoconference reference, and, since last year, offers LiveHelp and the Collaborative Digital Reference Service. LiveHelp uses e-commerce customer service software (LivePerson), requires only a network connection and Web browser, provides an organized database of canned answers, and includes the ability to push Web pages to users. Through LiveHelp, Cornell's remote users have access to a live, interactive reference service. Even students within viewing range of the physical reference desk have used LiveHelp rather than leave their workstation. Even more telling is Cornell's finding that undergraduate students want only electronic information and are not interested in using print reference materials.

The lessons learned at Cornell echo those learned at MCLS and the Santa Monica Public Library. Users expect digital services 24/7. Libraries must develop new models of service and redefine their definitions of primary users, including those who are on-site in relation to those who are not. According to Constantine, as librarians develop these new services, they should draw upon long-standing core values and not abandon traditional services. Librarians have to experiment and take risks, recognizing that there is a price tag with increased stress, frustration, and potential mistakes. Collaborative efforts, particularly with libraries in other time zones, can better serve users by extending times of availability. He closed by saying that the sign of the times is digital.
 

LC's CDRS
Having heard two speakers refer to LC's Collaborative Digital Reference Service without much explanation, we were ready for Diane Nester Kresh's in-depth presentation. After describing its rationale with trends about the short life span ofWeb sites, Internet use, and challenges that libraries are facing, Kresh stated the purpose of CDRS: to provide professional reference service to users anywhere, anytime, through an international, digital network of libraries. With this service, an end-user may ask a question and receive an answer from a librarian working through a participating member organization. The infrastructure supporting the system includes operating agreements describing the scope of services and software for tracking and managing the routing of questions and answers. LC is also building resource databases that will include member profiles and a searchable knowledge base of questions and answers.

Kresh reported that in November 2000, LC finished the final test of the CDRS's Phase 3, in which more than 50 members had participated. The goals of the pilot tests included developing a Web form for the question-and-answer process, creating procedures for assigning and tracking, measuring response time and interoperability, determining the scope of researcher requests, and defining best practices. Kresh and Linda Arret, LC's senior network specialist, gave a brief, live preview of CDRS. More information about CDRS is available at http://www.loc.gov/rr/digiref.
 

OCLC, LC Partnership
The capstone of the symposium was the announcement that the Library of Congress and OCLC have agreed to collaborate in designing and delivering a new reference service based on the CDRS pilot project. According to the agreement, OCLC will provide technical and developmental support by building and maintaining a database of profiles of participating institutions that will provide answers through CDRS; building and maintaining a question-and-answer database system that will enable CDRS participants to catalog answers and store them in a searchable/browsable database; and providing administrative support for CDRS, including marketing, registration, training, and user support.

Chip Nilges, OCLC's director of new-product planning, and Frank Hermes, OCLC's vice president of discovery and fulfillment services, presented the emerging paradigm that combines local, statewide, regional, national, and international member collaboration. The opportunities to be addressed include collaborative knowledge base systems, standards for question interchange, shared access to expertise, and standard metrics and accounting systems. OCLC views its role as one that supports emerging networks, partners with commercial service providers, delivers a low-cost alternative for local use, develops supporting services, and supports cooperative efforts to deliver services in the public Web space.

OCLC has an ambitious timeline for piloting system elements this spring, and expects to have a production version available around July. More information on OCLC's strategy is available at http://www.oclc.org/strategy. Most interesting to me was the goal to evolve WorldCat from a shared database to an open and global learning community. It's incumbent upon OCLC member libraries to collaborate and participate in the development of this system.
 

Conclusion
The Virtual Reference Desk is a timely concept, and this well-organized symposium offered substantive content, from vision to the trenches. The creativity, energy, humor, and persistence of the librarians who spoke balanced the formal corporate positions of LC and OCLC. Several speakers underscored the need for risk-taking at this juncture in the evolution of the virtual library. Their prophecy was that today's products will be deemed clunky a few years down the road. Digital reference is the development to watch in 2001.
 
 

Laverna Saunders is dean of library, instructional, and learning support at Salem State College in Massachusetts. Her e-mail address is saunders@noblenet.org.