The inter-linking of distributed electronic resources is one of the hallmarks of the Web. An area of linking now receiving a lot of attention is e-journals; the ability to link to e-journal articles from databases and from references in other e-journal articles is being implemented in many systems today. The most important development in this sphere is the Digital Object Identifier (DOI): an identifier that is also a "hot" link. Sixty-eight publishers of electronic journals have already joined the CrossRef organization, the DOI agency for e-journals. One difficulty (usually known as the "appropriate copy problem") with the DOI is that it currently only leads to a single copy of an article, usually the one in the publisher's own site. If an institution has arranged for access to the article through another source (an aggregator service, local loading, etc.), these alternate sources will not be found in the linking process. To address this issue, an experiment involving the International DOI Foundation, CrossRef, CNRI, Ex Libris, OhioLink, Los Alamos National Laboratory, Ohio State University, and the University of Illinois is underway to test a proposed model to allow the "localization" of the DOI link. The model integrates the DOI/CrossRef linking system and the OpenURL framework for open context-sensitive linking.

The principal use of the web today is human-access to documents and applications. Within a few years though, the web will need to support program-to-program communication, employing a variety of messaging protocols to support a broad range of applications -- from simple RPC requests, like "Get Stock Quote", or "Validate Credit Card", to complex session-oriented transactions, like search and retrieval. A common architecture or foundation will be necessary, to support these communicating applications; various approaches are being considered. There is a general assumption that the applications to be supported by this infrastructure will be based on XML protocols (that is, protocols whose message formats are described by XML and which use XML encoding). Some protocols that are described by other syntaxes may need to convert to XML to avail themselves of this infrastructure; an example is Z39.50. Conversely, some existing XML protocols are currently engineered to run over HTTP (not SOAP or XP). This session will examine this XML Protocol infrastructure, including SOAP and the emerging W3C XP protocol, and look at three protocols in particular: CIP, OAI, and Z39.50.

The ACLS History E-Book Project, with a $3 million grant from The Andrew Mellon Foundation, is planning to launch its first collection of e-books in Summer 2001. During the course of its first five years (1999-2004), the Project will publish a series of 85 new electronic books in history from ten university presses "University of California, Columbia, Harvard, Johns Hopkins, MIT, University of Michigan, New York University, University of North Carolina, Oxford, and Rutgers" and launch a series of 500 related backlist titles along with their major reviews. Five affiliated learned societies are participating in this project: the American Historical Association, the Middle Eastern Studies Association, the Organization of American Historians, the Renaissance Society of America, and the Society for the History of Technology. This session will provide a brief project overview, including technology, subscription, and publishing issues.  <http://www.historyebook.org/>

In this update session, IMLS will discuss:

• National Leadership Grant (NLG) projects of particular interest to CNI members
• Trends in both museum and library NLG applications for the next series of awards
• The 2001 Technology & Digitization Survey, which includes museums and libraries
• IMLS's legislative reauthorization, in progress
• IMLS's agenda: where the agency is going, how it intersects with CNI's constituents, and technology trends and research the agency is involved in

This presentation describes development of a model integrated storage, archival, and sketch-based query and retrieval system for 3D objects developed under an NSF KDI grant at Arizona State University. The initial focus has been 3D scans of Native American ceramic vessels, which have been defined as a set of three-dimensional triangulated meshes composed of points and triangles. This system models the data with parametric surfaces, extracts features of the vessel, and provides vessel measurements more accurately than possible using traditional tools of anthropology. The project uses a class-based XML schema to catalog and organize the 2D and 3D vessel data. A Web-accessible visual query process developed for the project permits users to search the databases of vessel data through sketches of sample vessel profiles in their browser window, or by selecting sample vessel shapes from the provided menus, in addition to traditional text and metric search criteria. The interface provides access to original and modeled data, and interactive 2D and 3D models. The ASU KDI project will be addressing additional data sets with different modeling requirements including lithic tools, condyle surfaces of bones, cellular DNA structures, 3D ultrasound data, and diatom shapes. The goal is to create an extendable model for a digital library of 3D data that captures, models, catalogs, searches, retrieves and permits interactive analysis of the data.

BECites+ is a pilot project of the Library of Congress Bibliographic Enrichment Advisory Team (BEAT). An outgrowth of another continuing BEAT Project, the Digital Table of Contents project in which the catalog records of selected monographs are enhanced with links to their scanned tables of contents, the BECites+ project is designed to test the feasibility of enhancing staff-produced bibliographies by not only placing them on the Web in electronic form, but also by adding links to scanned files of their tables of contents, indexes, and back-of-book bibliographies. In addition, reciprocal links are made between all of these data elements and to and from the online catalog record for each title in the bibliography as well as to the online guide in which the title is cited.

This cross-linkage results in enhanced information retrieval, as each of the links connects a searcher to other related resources and to an electronic bibliography on the same or similar theme. Further, this enhanced content information is available both to researchers who first encounter the citation through the bibliography on the Web as well as to those who come across the catalog record for one of the works cited in the "webliography" during a traditional search of the Library's OPAC. Finally, links to pertinent online journal indexes, other related Web resources, and to applicable subject headings in the Library's OPAC are also included. In addition, the project is experimenting in its business history guide with creating a master company name index that will direct researchers to those titles within a bibliography which contain information on a particular company.

An increasing number of campuses have joined the Internet2 project and the environment is ripe for partnerships between high performance networking providers and content providers. This session will focus on a discussion of strategies for libraries to become more involved with Internet2 projects and for developing campus partnerships between information technology units and the library. The recommendations of the ARL Internet2 committee will be discussed, and input solicited for additional means to promote library content and services as part of Internet2 projects.

Dartmouth College, in collaboration with Internet 2, has formed a PKI Lab to advance the state of the art in network services for authentication and authorization. The lab is investigating a variety of topics including interoperability and manageability, trust and delegation models, and scaling in the technical and human dimensions of PKI. Research at the Lab is exploring the interaction of two major aspects of the emerging distributed information world: trust and heterogeneity. For effective distributed information services, entities need to make trust judgments about other entities, in the context of massive variation in users, roles, computers, organizations, administrative domains, and application contexts. Public key cryptography is a uniquely appropriate tool for these issues, because it enables robust expression of non-trivial, compound statements and beliefs, among entities that share no common secrets. This session will provide some background on the structure and uses of Public Key Infrastructure in networked applications, identify some significant current problems and discuss research efforts underway to develop practical solutions to them. Some more background info can be seen @  <http://www.cs.dartmouth.edu/~pkilab/>.