Volume 9 Number 2
Digital Reference: An Overview
Director of Research and Development, Information Institute of Syracuse
Assistant Research Professor, The School for Information Studies
Digital reference refers to a network of expertise, human intermediation and resources placed at the disposal of users in an online environment. It employs automated tools wherever possible, allowing human experts to concentrate on "hard questions". But human expertise is expensive and hard to find. Automated tools are less expensive to incorporate into online services and sites, allowing digital libraries to use tools that were, until recently, the province of a small cadre of people.
For decades, professional searchers, information brokers and reference librarians have had access to powerful and precise search engines and other tools. Information professionals learned well how to wield these tools, but average web users have not had the benefit of similar training. In short, the Web has brought to the general public many tools, but not much of the expertise required to use them.
While we benefit from systems built on natural language processing, vector-based information retrieval and other helpful innovations, automated systems cannot yet rival a human being in the disambiguation of questions and synthesis of answers. In addition, although we aspire to digital library systems that are 100% intuitive and require no practice, assistance, understanding or study, such systems don't seem likely in the near future. So what happens when the user has a difficult question?
Until digital library tools can pass the Turing test , digital reference must incorporate people into automated systems to supply answers. Attendant to that circumstance are myriad issues, several of which are presented in this issue of D-Lib Magazine.
In "Current State of Digital Reference in Primary and Secondary Education", Lankes examines the domain of digital reference services in primary and secondary education. He argues that these services represent a revelatory case for digital reference, serve as a research environment, and provide models, theory and frameworks that will inform future research in the digital reference and digital library domains.
In "The Technological Challenges of Digital Reference: An Overview", Penka focuses on the need to understand the technical environment from the perspective of patrons and libraries, rather than simply adopting the newest technology.
In a related discussion, "Question Negotiation and the Technological Environment", Janes and Silverstein examine how automation can complement the ways in which people manage question negotiation, receiving, routing and answering.
White, Abels and Kaske report findings from a pilot study of chat reference in their article "Evaluation of Chat Reference Service Quality Pilot Study". Findings from the pilot study inform methodology, and discussion of average session length, queuing time, number of turns per session, down time, and lag time will be of interest to service quality researchers.
Goodrum's contribution, "Image Intermediation: Visual Resource Reference Services for Digital Libraries", points out that digital image intermediation is an increasingly important focus of research in library and information science. She notes, however, that users encounter difficulty with online tools for locating images, and she proposes to develop, test and evaluate procedures and mechanisms to support digital image intermediation. This research is designed to advance the digital reference research agenda in image intermediation, contribute to the creation of software and systems to support image intermediation, and guide the training and education of reference librarians and visual resource providers.
This issue of D-Lib is not intended solely for digital reference professionals. We hope all digital librarians who organize collections, design interfaces, and create services will consider these and related issues, enter the dialogue, and contribute to the theory and practice of digital reference.
 Turing, A. (1950), "Computing Machinery and Intelligence", Mind 59(236), pp. 433-460. Turing argued that if a human being and a computer were interrogated, and the results were indistinguishable, the computer might be considered to be "intelligent". Turing's work was one of the foundations of artificial intelligence.
Copyright © 2003 Joanne Silverstein
Editorial | Letters | Contents | First Article
Home | E-mail the Editor