Collaborative Project to Enhance Library Catalog Browsing

Robert Kieft, Haverford College

Introduction

The practice is so familiar that it seems to be governed by a law of nature: having found a potentially useful title by whatever means, library users typically go to the shelf and browse left and right in order to discover and select additional relevant titles. Powerful though the belief in serendipitous shelf-browsing discovery is, the efficacy of this "instinctive" practice is challenged today as libraries distribute increasing percentages of their printed collections to off-site storage or through collaborative archiving agreements. Under these conditions, users who rely on browsing the library shelf for the purposes of discovery and selection risk missing more and more material that might be of interest. Anecdotal and transaction log evidence has it that few use the browse feature of library online catalogs, not only because it is uninteresting visually but because the information users need in order to select what they want is not present. Until recently there has been no recourse except to the stacks.

Background

For almost three decades, librarians have advocated the enhancement of online library catalog records with such "evaluative content" as tables of contents, sample text, indexes, reviews, cover images, etc. in order to aid users in discovering and selecting, without leaving the library catalog, the materials they might want to read. To this end, many libraries have followed the lead of online booksellers by adding evaluative content for current publications to their catalogs. Such content is not commercially available, however, for pre-1990s or most foreign publications, that is, those to which libraries are most likely to offer remote access.

Initial activities

At ALA Midwinter 2004, a group convened by Bob Kieft, Librarian of the College, Haverford College, and Susan Perry of the Mellon Foundation and CLIR, with the support of David Seaman of the Digital Library Federation, met to discuss the establishment of a cooperative effort among libraries to create and share indexable evaluative content for older materials. The meeting involved representatives from the Library of Congress Cataloging Directorate (John Byrum and John Celli), the national bibliographic utilities (Judith Bush, RLG, and Deb Bendig, OCLC), the bookselling community (Cindy Cunningham, then of Amazon.com, Sam Dempsey and Bob Bogan, Baker and Taylor), NISO (Sally McCallum), and libraries (Karen Schmidt, University of Illinois, and Roy Tennant, California Digital Library). Since the meeting, Bernard Reilly, President of the Center for Research Libraries and convener of the July 2003 conference "Preserving America's Print Resources" (PAPR), and Dan Hazen, Harvard University, have joined the discussion.

To help refine the project and recruit collaborators, Roy Tennant, Deb Bendig, Sally McCallum, and Merrilee Proffitt (RLG) will host a discussion at the April DLF Forum in New Orleans. Bob Kieft, Karen Schmidt, and Dan Hazen will organize an open meeting at ALA Annual in Orlando for the same purpose. Duane Webster of ARL has agreed to present the project at the May meeting of ARL's Collection Development Committee.

Given the success of booksellers in moving browsing online, the conditioning of user expectations by the Web, and the availability of proved technologies, the members of this ad hoc group feel that now is a propitious time for libraries to create and share evaluative content for the millions of items in their collection. They propose to develop a program that will enable distributed, non-duplicative input of evaluative content using inexpensive staff and workflows and an initially simple, but potentially richer and more complex data packaging structure created from such emerging standards as METS, MODS, OAI-PMH, and ONIX.

Having these augmented records in library catalogs will remove many of the barriers customarily encountered in resource-sharing and remote storage programs by putting libraries in a position to persuade their users that they can better fulfill their discovery and selection needs in the catalog than they can at the shelf. The group plan to have a full technical description of the record structure and record creation protocols ready later in 2004 and a demonstration project running soon thereafter.

From: CLIR Issues (March/April 2004)