Digital Library Federation. Report from the Office of the Director for the period October 2000 to June 2001
Submitted for information to the DLF Steering Committee at its meeting of 19 June 2001.
From inception, the DLF set itself the task of "creating the means to bring together digital materials - from across the nation and beyond - that will be made accessible to students, scholars, citizens everywhere". As it enters its sixth year, it is settling on a process for achieving this ambitious aim. In particular, the DLF is demonstrating its ability in:
Priorities for DLF investments are determined with respect to members' articulated needs and to the opportunities that are created by or presented to the DLF for responding to those needs in a meaningful way.
DLF funding stimulates the development of so-called infrastructural or utility tools and services that are commonly required by digital libraries but beyond their independent means. Work in this area develops functional specifications and business cases for the development of these tools or services. Working in this way, the DLF seeks to reduce the risk to investors who have the financial capacity to develop such tools and services. In this respect, DLF investment may be seen as incentive or venture funding.
2.1. Academic Image Cooperative (AIC).The DLF's work prototyping organizational, business, and technical aspects of an image distribution service contributed directly to the formation and launch recently by the Mellon Foundation of ArtSTOR - "an independent not-for-profit organization that will develop, 'store', and distribute electronically digital images and related scholarly materials for the study of art, architecture, and other fields in the humanities". ArtSTOR marks a major advance in the development and dissemination of visual image resources that support research and teaching. The DLF owes a debt of thanks to Max Marmor, Distinguished DLF Fellow, for guiding this work.
2.2. Shared cataloguing tool for visual resources.ArtStor has also subsumed the DLF's work in this area to develop a tool to minimize the vast redundant effort involved in the development of high-quality visual resource descriptions. Beginning in January 2001 under the guidance of DLF Distinguished Fellow Max, Marmor, the DLF developed an outline functional specification and market assessment for the cataloguing tool.
2.3. Open Archives Initiative (OAI).Through its investment in the OAI, the DLF has helped to supply a network protocol upon which the next generation of scholarly Internet portal services is likely to be built. The OAI is being taken up enthusiastically in the US and abroad and Internet services that use the OAI are attracting significant investment from The Andrew W. Mellon Foundation, the National Science Foundation, and the European Union Directorate General XIII. In addition, the DLF is playing a role facilitating the development of harvesting services. To date, twelve DLF members have agreed to contribute to new Internet portal services, the metadata from over 50 collections (representing some millions of objects). Metadata from several of these collections are already available through OAI conformant servers that have been developed at the Library of Congress, the University of Illinois at Urbana Champagne, the University of Pennsylvania, and the University of Tennessee at Knoxville. In addition, five DLF members and one DLF ally have been invited by The Andrew W. Mellon Foundation to submit proposals to develop harvesting services built in part on these metadata. Funding decisions pertaining to these proposals will be taken in mid-June and should be available for report at the Steering Committee meeting.
2.4. Digital preservation.Building on work conducted by CLIR, DLF, and CNI on the minimum criteria that libraries and publishers may require of an e-journals repository, The Andrew W. Mellon Foundation established its e-journals archiving program and funded seven institutions to plan the development of such repositories. With this program, libraries and publishers take a significant practical step forward in addressing their shared preservation concerns. The DLF will support the program by hosting its web pages, reporting its progress to the broader community, and encouraging cross-fertilization amongst its funded participants. Six of the seven institutions funded by The Andrew W. Mellon Foundation are DLF members (Cornell University, Harvard University, the New York Public Library, the University of Pennsylvania, Stanford University, Yale University).
2.5. Reference linking.With NISO and other organizations, the DLF has been successful in encouraging CrossRef to explore a reference linking architecture that will not constrain the choice that libraries currently exercise when selecting e-journal content providers.
2.6. Registry for digitized books and serials.The DLF is developing the functional specification and organizational and business cases for a service that records information about the existence and accessibility of digital surrogates and about the existence, location and locus of preservation responsibility for those surrogates and for the artifacts from which they were made. Such a service would provide numerous benefits to libraries including the substantial reduction in their costly redundant effort e.g. in digitizing and in managing digital and print copies of books and serials.
Leveraging its members' collective influence the DLF identifies, endorses, and promotes those standards and good practices that support the digital library's interests in high-quality, interoperable, persistent, and sustainable online collections and services. It is also developing the network of strategic alliances it will need to ensure that its good and best practice recommendations are maintained and communicated to the broadest possible community.
Although there is an opportunistic element in the way that priorities are assigned to work in this area, effort increasingly grows out of the best practice requirements that surface in developing functional and business cases for essential utility tools and services (see Section 2 above).
3.1. Liblicense model agreement.After review, the DLF has endorsed this model license agreement for use by libraries and commercial publishers. The model license effectively documents preferred and good practice and acts as a decision tool that is likely to save libraries both time and money in negotiation robust contracts with commercial content providers. Ann Okerson (Yale University) will maintain the model license so that it evolves to reflect new circumstances as they arise.
3.2. Metadata encoding and transmission standard (METS).The DLF has initiated a process to identify standard structural, administrative and technical metadata for digital objects. Such a standard is an essential pre-requisite the construction of reliable and persistent distributed digital library collections that may be available through harvesting or registry services as described in 2.3. and 2.6., respectively.
3.3. Benchmarks for digitally reformatting printed books and serials.The DLF has initiated a process to develop consensus within the community about benchmarks standards for digitized books and serials; that is, benchmarks that define when a digitized book or serial:
3.4. Methods for evaluating digital image quality.Building in part on work on imaging practices conducted with the Research Libraries Group (RLG), the DLF is investigating methods for evaluating the quality of digital images. Such methods will help assess claims that made about image (e.g. that they meet specified benchmarks).
The DLF pools research and development expertise at member and other institutions to survey and evaluate existing practice and to answer pressing strategic, organizational, and technical questions. Work in this area is driven by members' perceived needs, by opportunities that are available to conduct meaningful research, and, increasingly, by the questions arising from DLF investigations into infrastructural tools and services (see Section 2 above)
4.1. Strategies for developing sustainable digital collections.Based on survey of practice at leading research libraries, the studies recommend strategies for developing collections from commercially supplied electronic content (Tim Jewell, University of Washington), digitally reformatted content (Abby Smith, CLIR), and links to third-party public domain Internet content (Lou Pitschmann, University of Wisconsin). The studies (available now as online drafts and soon in print) provide important decision tools for those developing digital collections.
4.2. Digital library policies, organizations, and practices.The DLF has completed a member survey designed to identify the institutional contexts in which digital libraries are being developed. By documenting the very different paths along which digital libraries evolve, and the different ways they organize and fund themselves, the survey will: inform strategic planning and decision making within digital libraries; provide benchmarks for assessing digital library development; and identify emerging library roles. A draft report based on the survey data is available from the DLF website. A formal report that will contain case studies that document the key digital library development paths that are apparent in the data will be available at the end of the calendar year.
4.3. How and why library use is changing.Assisted by DLF Distinguished Fellow Denise Troll, the DLF convened directors from eight academic libraries to explore how we might enhance our understanding of changing patterns of library use. Chosen to represent a range of academic institutions, the library directors agreed that DLF should work with Outsell Inc. to examine how faculty, students, and staff at very different academic institutions, perceive of, locate, and use scholarly information and information services. The study, when conducted, will provide data that will help us to understand how academic libraries fit in a broader information landscape and as such environmental data that is essential to the interpretation of library use statistics. A draft proposal for the study has been prepared and funding is presently being sought.
4.4. Methods for assessing use of online collections and services.DLF Distinguished Fellow Denise Troll is completing a survey of practices at DLF member institutions. Based upon 65 interviews, the study promises an assessment of current practice.
4.5. Issues in electronic scholarly publishing.In May 2001, the DLF convened a small group of representatives from university libraries and university presses to discuss their different approaches to electronic scholarly publishing, to establish a venue for information sharing about appropriate tools and technologies, and to encourage collaborative developments. The meeting is an initial first step in an exploration of production and distribution needs that may be common to libraries and university presses and as such met through some common solutions.
To build a community of professionals, the DLF facilitates its members' communication, collaboration, and cross-fertilization. Early indications from an evaluation instrument recently circulated to staff at member institutions suggest the DLF's newsletter (now published twice rather than four times a year) and its extensive and growing website are well received.
The DLF forum meanwhile has evolved as both a major success and a leading digital library conference. Open to staff at DLF member institution, the bi-annual three-day event attracts between 100 and 125 people to meetings that provide opportunities for information exchange, professional development, and participation in and assessment of DLF initiatives.
Evidence of the DLF's community-building progress is evident in the fact that the majority of its new initiatives now originate outside the Director's Office in informal discussions amongst professionals at member institutions.
Evidence that DLF initiatives are having an impact on the broader community is available meantime in the growing (in fact, unmanageable) request for information about DLF initiatives and their significance for non-member libraries and other organizations. The DLF is currently assessing how best to respond to this demand.
DLF Membership. The DLF extends a warm welcome to its newest member, the University of Washington. Lisbeth Wilson, will represent the University on the DLF Steering Committee.
DLF evaluation. The DLF was established with "a sunset clause" and its funding runs to June 2002. To inform the members' discussion of possible futures, the Director's office is supporting a formal evaluation. That evaluation, to be conducted by directors of DLF member institutions will be a summative one that assesses the impacts the DLF has had since its inception and the costs and benefits of its various initiatives. It will also be formative and explore possible developmental trajectories for the DLF beyond June 2002. The evaluation is expected to take place over the summer months and conclude at the November 2001 meeting of the DLF Steering Committee.