The NSF - DARPA - NASA Digital Libraries Initiative - Mid-Term Perspectives by Stephen M. Griffin The Digital Libraries Initiative (DLI) projects sponsored jointly by the National Science Foundation (NSF), the Department of Defense Advanced Research Projects Agency (DARPA), and the National Aeronautics and Space Administration (NASA) are now mid-way through the initial four-year funding period. This article focuses on the success of these projects in building partnerships with other organizations and influencing the broader digital libraries community. The six DLI projects function and are managed as a collaborative effort with the common goal to dramatically advance the means to collect, store, organize and use widely distributed knowledge resources containing diverse types of information stored in a variety of electronic forms. Each of the DLI projects has a research and testbed component and brings together multidisciplinary teams of researchers and users from a lead university with those from partnering organizations. Partners include major US computer and communications com-panies, academic institutions at all levels, libraries, publishers, government and state agencies, professional associations, and other organizations with a stake in the development and use of digital libraries. The projects are centered at:
University of California, Berkeley University of California, Santa Barbara (http://alexandria.sdc.ucsb.edu/) University of Illinois, Urbana-Champaign (http://www.grainger.uiuc.edu/) University of Michigan, Ann Arbor (http://www.sils.umich.edu/UMDL/HomePage.html) Stanford University Carnegie Mellon University (http://informedia.cs.cmu.edu/) DLI Projects and Digital Libraries Research Research in digital libraries explores the development and use of large-scale networked knowledge repositories. The goals are to efficiently and effectively capture, store, search and retrieve knowledge from distributed electronic collections. The collections might contain text, images, maps, audio recordings, video and film clips and combinations of these (multimedia). The DLI projects are exploring a broad range of cutting-edge research topics including semantic interoperability, image retrieval by content, new models for distributed documents, strategies for rights management, network economics and billing and social aspects of digital libraries. Partnering The DLI projects have been exceptionally successful in building partnerships. These have resulted in technology exchange, transfer of knowledge and resources, collaborative research, and valuable personal interaction. Through partnerships, the DLI projects have been able to enlarge their topical scope, increase staffing, expand testbed content and functionality and gain additional support for basic research. Individual projects report more than fifty persons directly participating in research activities and cost-sharing exceeding two hundred percent of their base Federal funding. Partner relationships differ in degree and in type, change over time and involve differing levels of interaction. For the projects a partner might:
contribute to funding base technologies and infrastructure provide access to advanced technologies send researchers to work on-site at the university contribute content for the testbed and/or information services. In return the projects involve partners in the experiences of their effort, making them early users of research products and offering opportunities for intellectual enrichment. A short list of major DLI technologies partners includes Xerox, IBM, Hewlett-Packard, Digital Equipment, Sun, Ricoh, Hitachi, Bell Atlantic, Intel, Illustra, Oracle, SoftQuad, OpenText, et.al. Content providers include Elsevier, John Wiley, MIT Press, Open University, Telecom Italia, numerous professional societies and federal and state agencies. Outreach and Community-level Activities An essential component of the Digital Libraries Initiative are those activities designed to engage the broader community. D-Lib Magazine, coordinated by the Corporation for National Research Initiatives (CNRI) and funded by the Defense Advanced Research Projects Agency (DARPA) is an example. The WWW home page is http://www.dlib.org. The DLI has supported a number of workshops on technical topics: IITA Workshop on Digital Libraries. (http://www.hpcc.gov/reports/reports-nco/iita-dlw/main.html) UCLA-NSF Social Aspects of Digital Libraries Workshop. (http://www.gslis.ucla.edu/DL/) Thirty-Seventh Allerton Institute Workshop (http://edfu.lis.uiuc.edu/allerton/95/) DLI Metadata Workshop (http://alexandria.sdc.ucsb.edu/metadata/) SGML Mathematics Workshop Future Programmatic Directions Based on experiences of the first two years of the DLI, numerous topics have emerged as prime areas for expanded effort and resources. The list continues to evolve both in terms of membership and ranking of importance. An overall observation is that as projects mature and digital libraries research advances, views on digital libraries design and assessment are expanding from strictly techno-centric to include human-centric perspectives. Digital libraries are one manifestation of the human-centered information systems of the future. A partial list of promising topics would include:
new information objects, knowledge representations, integration schemes domain-specific repositories processes of human understanding of information from content, structure and context intelligent agents on networks interoperability metadata, including terms and conditions rights management of intellectual property including naming and secure object store new forms of collaboration and social informatics. To this list must be added topics of particular concern to the international community: foreign language query and search, OCR, navigation and multilingual translation tools for effective and timely access to global knowledge bases. Agency program managers meet regularly to discuss current developments and consider future directions. Several guiding principles are observed in contemplating the structure and scope of future activities:
planning processes should be pluralistic, with broad representation and community involvement future digital libraries research is best considered within a large analytical framework which includes knowledge, information, systems, users, and communities in the broad sense of the term topical considerations should shape individual projects and programs, not the reverse agencies should look to alternative research modalities for collaborative development and use of digital libraries. Conclusion New digital libraries capabilities point to high future demand for new services throughout the world. Network-based knowledge repositories promise to revolutionize many aspects of modern societies. The Digital Libraries Initiative will contribute to these exciting transformations by generating new knowledge, promoting innovative thinking, and accelerating the technology transfer process across industrial, commercial, scholarly, educational and social dimensions of societies. Ultimately, these efforts will help make vast amounts of useful information easily available to large numbers of people with diverse information needs. The material in this article represents the views of the author and does not necessarily reflect National Science Foundation policies. For more information: http://www.cise.nsf.gov/cise/IRIS/IRISHome.html Please contact: Stephen M. Griffin - Division of Information, Robotics and Intelligent Systems Program Manager: Digital Libraries Initiative Tel: +1 703 306 1930 E-mail: sgriffin@nsf.gov
