The Open Archives Initiative and Digital Libraries

These pages introduce the OAI and supply a brief history of the DLF's involvement with it. The DLF is also evaluating the OAI's technical framework by developing a number of Internet gateways that integrate access to distributed digital library collections. An account of those evaluation projects is available by clicking here.

Introduction

The OAI has developed a technical framework for facilitating the efficient dissemination of content via the network. One very obvious use is in the development of Internet gateways through which users can search simultaneously across diverse and geographically distributed online catalogues, finding aids, and other metadata databases.

The OAI framework is based on an approach known as metadata harvesting. In this approach, there are data providers and service providers. Data providers (such as individual libraries, museums, archives, data services, and e-print archives) support a simple harvesting protocol to provide extracts of local metadata in a common, minimal-level format in response to requests from service providers. Service providers use extracted metadata to build higher level, user-oriented services, such as catalogs and portals to materials that are distributed across multiple libraries, museums, archives, and other repositories.

History

The OAI has its genesis in an October 1999 meeting held in Santa Fe, New Mexico, under the sponsorship of the Council on Library and Information Resources (CLIR), the DLF, the Scholarly Publishing and Academic Resources Coalition, the Association of Research Libraries, and the Research Library of the Los Alamos National Laboratory. Focusing on the interoperation of "e-print archives" (collections of electronic journal articles and preprints), the OAi developed the harvesting approach and its protocols and documented these in the "Santa Fe Convention," along with preliminary ideas about acceptable use policies, registries, and other issues.

Almost immediately upon its publication, the Santa Fe Convention attracted considerable interest from libraries, publishers, museums associations, and other bodies who saw significant potential benefits in metadata harvesting. A meeting of the Digital Library Federation convened by Harvard University in May 2000 with support from The Andrew W. Mellon Foundation, for example, expressed its hope that the Santa Fe Convention would support the development of Internet gateways or portal services through which users could access information about library holdings irrespective of their location and format. This interest is expressed in a vision statement prepared after the Harvard meetings.

The second OAI workshop, held in conjunction with the ACM Digital Libraries meeting in San Antonio in June 2000, reviewed early implementation experience with the Santa Fe Convention and the interests in generalizing it to support the needs of a broader community of information providers. Out of that meeting came a consensus that the Santa Fe Convention needed to be revised and updated with the intent of producing a new more general version by January 2001. The meeting also accepted the need for some very limited organizational support for the OAI which had hitherto run entirely on the voluntary effort of a small number of dedicated individuals.

In August an OAI Steering Committee was established to oversee the OAI's development. The Committee convened a technical working group that met in September 2000 to revise the Santa Fe Convention. It also agreed that the convention, once revised, should be fixed for a period of one or two years, and that any further technical work be based on any implementation experience gained during that time. The Committee finally agreed to establish some organizational support for the initiative, and with funding from the DLF and the Coalition for Networked Information, that support was located at the Computing Science Department of Cornell University.

At present, the OAI has revised its technical framework (renamed the OAI harvesting protocol) and is encouraging practical evaluation, notably in the development of prototype harvesting services. The DLF is actively involved in this evaluation activity and is focusing its efforts on the development of services built with metadata that are harvested primarily from library systems. More information about the DLF evaluation projects is availably by clicking here.