1. Digital Library Repositories and Instructional Support Systems: Repository Interoperability Working Group Leslie Johnston University of Virginia Library Coalition for Networked Information, 4/16/2004

2. Repository Interoperability Working Group People: –John Mark Ockerbloom, University of Pennsylvania –Leslie Johnston, University of Virginia –MacKenzie Smith, MIT –William Ying, ArtSTOR Goal: Creation of a Checklist that outlines both essential and desirable interoperability functionality between Repositories and Learning Applications.

3. Checklist Assumptions: Use of Repositories A scholar is likely to require materials drawn from multiple repositories to support research or teaching. Teaching is increasingly supported by learning applications. To make the most effective use of digital content in teaching, learning applications need to be able to easily interoperate with digital repositories.

4. The Checklist focuses on the flow of information from repository to user rather than on content deposit. The Checklist is intended primarily for those developing repository systems of any type and learning applications intended to work with them. –Metadata only, available digital media files, publications/pre-press, or learning objects within a learning application. Academic, not-for-profit, and commercially-run repositories will benefit from the Checklist. Checklist Assumptions: Scope and Audience

5. Scholarship is a three-stage process: –“Gathering,” where content is discovered, evaluated, and acquired for use. –“Creating,” where content is adapted for instructional use, or new content is created based on the information in the gathered content. –“Sharing” where the new or adapted content is then made available to others. The Checklist focuses on Gathering, where content is drawn from digital repositories, but the needs of the later stages are important for understanding necessary repository services. Checklist Assumptions: Scope and Audience

6. Tasks in Gathering –Discover sources of potentially useful content. –Search for content that meets their needs. –Collect references to relevant items they find. –Import items, descriptions of those items, or references to those items into learning applications. –Save copies of some of these items to local applications or storage. –Find Related items to those they have collected, in the same or different repositories. Of the activities above, the essential activities that a digital repository must directly support are searching, collection, and import. Checklist Assumptions: Scholarly Process

7. The Checklist assumes a simple data model: –There exist distinct, identifiable pieces of digital content, or items, that can be searched for, collected, and imported for instructional purposes. –Items can be found by searching collections, groupings of items that can be addressed and queried through a common interface or set of services. –Items, and possibly collections as well, have metadata associated with them, information that describes them and otherwise aids in their use and management. Checklist Assumptions: Data Model

8. Repositories provide content, not simply metadata, to users. Teachers and learners use the content of digital repositories through learning applications. –Courseware packages, citation managers, and presentation and analysis software. With the wide selection and range of repository interfaces, there has also emerged a layer of mediators between repositories and applications, or gateways, that help users locate content they need in appropriate repositories. Checklist Assumptions: Architecture

9. Component Architecture Diagram

10. Ensure broad accessibility of the repository. Provide access controls to items in a way that does not hinder learning applications. Checklist Assumptions: General Design Principles

11. Each section includes: –Rating of Essential or Desirable for each category of interoperability functionality. –Description of the functionality and what purpose it serves. –The place in the architecture of a repository. –Some technical recommendations, but not meant to be proscriptive. The Checklist

12. Finding content: –Support search for items. (ESSENTIAL) –Provide standard or documented metadata for items. (ESSENTIAL) –Support search via software agents. (DESIRABLE) The Checklist

13. Collecting content: –Provide stable references to items. (ESSENTIAL) –Support citations in recognized scholarly formats for items. (DESIRABLE) The Checklist

14. Accessing content: –Provide ways to get and use item content. (ESENTIAL) –Provide views of item content. (DESIRABLE) –Allow items to be copied into local systems. (DESIRABLE) The Checklist

15. Documentation: –Document policies and functions of the repository. (ESSENTIAL) –Make the repository, and its content, known to other applications. (DESIRABLE) –Document the technical profile of the repository. (DESIRABLE) The Checklist

16. Creation of a questionnaire spreadsheet based on the Checklist. –Representatives of six repository projects were asked to fill in the questionnaire. –Answers were reviewed to judge both compliance of the project to the proposed interoperability standards and to gauge the successfulness of the questionnaire in determining compliance. Feedback

17. The questionnaire answers produced a number of issues. –The questionnaire was not deemed fully successful, especially for larger environments where some portions of the infrastructure might comply and others do not. –Who answers the questionnaire changes the compliance. –How can the questionnaire take into account the necessary interoperability differences between repositories and catalogs, which are metadata repositories? A revised questionnaire will be created. Feedback

18. Use Cases: –The use cases developed by the other working group led to a number of changes in the Checklist, especially in the definition of the process of scholarly research, the delineation of activities that make up the "Gathering" stage of research, and the categorization of functionality as essential or desirable. Feedback