May 01, 2003
Readers will not be dazzled by glitzy graphics as they enter the Open Archives Initiative (OAI) Web site, but its potential impact on learning is greater than that of most glossy online brochures. As stated on the site, the Initiative "develops and promotes interoperability standards that aim to facilitate the efficient dissemination of content." In other words, it details a mechanism that can be used by colleges or individual professors to share their work freely over the Internet.
In practice, this involves two separate groups of resources. The first of these is the Protocol for Metadata Harvesting (OAI-PMH), which specifies how metadata that describe online resources can be gathered from repositories that support the protocol. The second is a set of tools that help people build their own OAI-compliant archives or harvesters.
Site visitors can find the OAI-PMH by clicking on the Documents link from the front page. Within that section is a short set of links, including access to the most recent version of the protocol. This document is heavy reading, however, and is intended for people who plan to write software. General readers will want to click on OAI-Related Papers and peruse some of the overviews provided there. Taken together, these overviews indicate how open archives offer new ways for academic publishers, museums, libraries, businesses, and other institutions to store, disseminate, and access information. A recent article by Jeffrey Young (2002), for example, explores the effect that open archives may have on traditional academic publishing.
Once grounded with an overview of the project, it is worth returning to the Documents page to read some of the more detailed descriptions. In particular, the paper entitled "The Open Archives Initiative: Building a Low-barrier Interoperability Framework" (Lagoze & Van de Sompel, 2003) is an authoritative and comprehensive description of the project that should satisfy the needs of most readers.
For persons interested in creating their own archives, the Tools section is required reading; the link to this section is accessible from the menu bar at the top of any page. The OAI Repository Explorer can be used to get an initial understanding of how a harvester—a tool that gathers resources from remote sites—interacts with an OAI-compliant repository. The Explorer is very popular with designers, who use it to debug their repository implementations. After becoming familiar with the system, readers can click on the OAI-Specific Tools link to view a selection of software available for downloading. Each tool is listed with a brief description its capabilities and the name of a prominent institution that helped develop it.
The OAI Repository Explorer will be confusing at first. It is designed not as a user interface, but as a demonstration of the different inquiries that a harvester can make of an OAI repository. It can query the repository for supporting information such as the identity of the repository, the formats that the repository supports to make metadata available, and the subdivision of the repository into so-called "sets." It can also issue requests that return metadata about resources. The most important criteria used in such a request is the date of addition or update of the metadata in the repository; this sort of request allows for incremental harvesting and the establishment of a permanent synchronization between the harvester and the repository. Harvesters typically use the returned metadata to provide value-added services across various digital collections.
Most of the tools listed by the OAI are designed to be run on a Web server; these are not browsers that users would download and install on their desktops like a Web browser. Included are archive and harvesting tools written in computer languages such as Perl, PHP, and Java. The tools that I tested installed very smoothly, but installation is recommended for Web technologists only. With the use of a harvesting tool, a library or resource Web site could enable readers to search through a large number of online repositories with a single request. The Arc source tool, for example, is particularly valuable because it combines a harvester with a supplementary tool that allows users to perform intensive searches across specific sets of harvested metadata.
For more information about the Open Archives Initiative, visitors should click on the Community link on the top menu bar. This page offers subscriptions to OAI mailing lists, allows people to register as data providers, and contains links to a number of OAI-member Web sites.
It is worth reiterating here that this is not a site intended for the casual reader. Though introductory material is available on the OAI Related Papers page, much of the rest of the content is technical and detailed; developers and implementers will find it more helpful than the average visitor. Yet the OAI is an important Spotlight Site selection because it represents the beginning of a new movement in scholarly publication. As individual and institutional archives become more common (as evidenced by the recent publicity surrounding DSpace, OAI-compliant repository software from the Massachusetts Institute of Technology), the need for more traditional forms of publication, such as print journals, will diminish. Additionally, because most content available through OAI repositories is free, the economics of journal publication will change.
Traditional journals do provide the valuable services of editing, filtering, and refereeing. These will remain important functions, regardless of whether publication is online or in print. But the major role of conventional publishers—printing and distributing copies of articles—will be largely subsumed by open archives like those envisioned by the OAI. Its Web site gives Technology Source readers a preview of the changes to come.
Lagoze, C., & Van de Sompel, H. (2003). The Open Archives Initiative: Building a low-barrier interoperability framework. Retrieved April 30, 2003, from http://www.openarchives.org/documents/oai.pdf
Young, J. R. (2002, July 5). "Superarchives" could hold all scholarly output [Information technology]. The Chronicle of Higher Education, p. A29. Retrieved April 30, 2003, from http://chronicle.com/free/v48/i43/43a02901.htm