About OAI targets
If your institution utilizes a web scale discovery or federated search system, we encourage you to expose your content using an OAI publishing target. An OAI publishing target exposes your content on our OAI server, which a discovery vendor may query for harvesting.
As the project administrator, you select the items that are ready to expose by publishing. Once published, any future updates to the item in JSTOR Digital Stewardship Services (metadata or media files) are automatically pushed to the OAI server with a ‘last update date’, allowing the vendor to come to the server on a regular basis to pick up any new or changed items. The data output of these records are in XML, in which you may choose to map the fields to Dublin Core, or leave the fields unmapped and thereby use the SSIO (System-Supplied Input/Output) schema. Mapping to a standard like Dublin Core is typically recommended for working with discovery vendors because it will simplify the harvesting and transformation process.
Establishing an OAI publishing target requires IP address authentication and Support assistance. For discovery vendors harvesting content from multiple institutions, the harvester will see all projects from all institutions in the same list when running the query request. Each project returned in the query request will be indicated by Project ID and Project Name. JSTOR Stewardship will provide the list of project IDs and project names that belong to each institution authenticated by that vendor’s IP address, so the vendor may write the harvester accordingly.
Setting up an OAI Target
To set up an OAI target for your project, please contact JSTOR Stewardship Support and include the following information:
- Project name
- Project ID (can be found at the end of the URL in your browser address bar when viewing the project)
- Discovery Vendor (or institution that will be harvesting the OAI data)
- Discovery vendor’s IP (or IP address of the institution harvesting the data)
Managing OAI field mapping
Unlike other publishing targets, fields from your project do not need to be mapped in order to display in an OAI harvest. The harvester pulls the data in XML format to be displayed in the discovery service as decided between you and the vendor. By mapping your fields, you establish the preferred XML data output. If you choose to map your fields, the OAI publishing target uses the Dublin Core schema. When mapping fields to the Dublin Core schema, use of prefixes, displayable, searchable, and required fields is not supported.