Principal Investigator: Adam Chandler
Central Technical Services
110C Olin Library
Cornell University
Ithaca, NY 14853-5301

Samuel Lazerow Fellowship
Association of College and Research Libraries
American Library Association

1 December 2000
Amount requested: $1,000

The focus of this research is the creation and management of metadata for licensed electronic resources. This project continues the 1999 Cornell investigation of the OCLC Cooperative Online Resource Catalog (CORC) system. At the conclusion of that effort a final report was written describing the results of Cornell’s trial (Calhoun, 1999). One of the conclusions in the report is an observation that the CORC system lacks functionality for supporting metadata required to manage acquisitions activities (see attachment, p. 13). What is needed in addition to what CORC offers, they wrote, is a system for managing local administrative metadata relevant to creating or maintaining access to a licensed resource; that is, metadata for managing a continuous library acquisitions process.

Managing electronic resources is complicated, involving many different pieces of often dynamic data. Researchers here at Cornell listed some 35 elements relevant to the management of licensed electronic resources in the Cornell environment (Calhoun, 1999). (A few of the elements are: selector, genre, access, number of simultaneous users, price, and payment history.) Tim Jewell at the University of Washington is involved in creating his own list of elements for managing electronic resources, based on surveys he is doing with other libraries (Jewell, 2000). His latest draft contains about 40 elements. Sharon Farb and a committee at UCLA are starting an investigation of the same problem (Farb, 2000). Carole Richter and her team at Notre Dame have already selected elements and created a working system, a web database, which they are using and planning to enhance (University Libraries of Notre Dame, 2000). It is worth noting that the Notre Dame system also contains nearly the same number of elements, about 45, an indication that a common understanding of the problem is emerging. Another ongoing effort, called ERLIC (Electronic Resources Licensing & Information Center), involves staff at Penn State University (Stanley and Holden, 1999).

Given the excitement and energy surrounding these questions, there clearly is an opportunity at the present time to corral the various individual initiatives into a standard solution, a solution less institution specific and thus more valuable to librarianship. We are at a critical point where fruitful cooperation is possible because interest in the problem is high, but not too much has been invested in installed solutions. What is needed at this juncture is a bridge between the various local efforts mentioned above and the international metadata initiatives. To be specific, we would all benefit by creating a bridge to the XML RDF standard.

One of the products of the Dublin Core Metadata Initiative is the Resource Description Framework (World Wide Web Consortium, 2000). The purpose of the XML RDF model is to allow the mixing of elements from different metadata vocabularies. The RDF model is well suited to this problem. Instead of creating different named data elements in each local system it is more sensible to borrow what we can from Dublin Core (title and date are two obvious examples), then within the interested library community agree on the rest of the elements needed to adequately describe and manage licensed electronic resources. This decision may then be made available to others by registering the element definitions with the DESIRE Metadata Registry (UK Office for Library and Information Networking, 1999). The next step in the process is to register the "application profile", that is, register the whole set of Dublin Core elements plus our domain specific elements chosen to solve the metadata needs of this particular domain (Heery and Patel, 2000).

After those decisions are made, work may begin on creating a general software solution, one which incorporates the standardized elements into the design. This solution is envisaged as a technically unremarkable Web-based database, a system adaptable to local library workflow. When we are confident that the software is reliable through testing at Cornell we will share the code with other interested libraries. All phases of the project will be completed before the end of 2001.

1. Identify and describe existing projects relevant to the goals of this initiative. Make contact with responsible individuals.

2. Create and maintain a Web site which will act as a central place for information about this research topic.

3. Identify a set of common metadata elements which support collection and management of electronic resources. This will be accomplished through a variety of means, including email, telephone, survey and site visits. We will then solicit more comments on the completeness of the proposed set of elements.

4. Develop a model for managing these metadata that fits the needs of the library community.

5. Register the final version of the licensed electronic resources metadata vocabulary and application profile with the UKOLN DESIRE Metadata Registry.

6. Develop and implement at Cornell a scalable software prototype for managing electronic resources based on our metadata application profile.

7. Share the research conclusions and software with other interested libraries.

An Application Profile and Prototype Metadata Management System
for Licensed Electronic Resources

Calhoun, Karen, et al. "CORC at Cornell Project: Final Report: Appendix 1: Data Fields to Support Selection and Acquisitions Activities." 11 December 1999. <http://campusgw.library.cornell.edu/corc/>

Farb, Sharon. "Administrative Metadata for Licensed E-Resources." E-mail to Adam Chandler. 27 November 2000.

Jewell, Tim. "Administrative Metadata for Licensed E-Resources." E-mail to Karen Calhoun. 22 November 2000.