Rdf Primer / quote 25 of 55
PRISM: discovery, rights tracking and end-to-end metadata
This overall goal requires a metadata approach that emphasizes discovery, rights tracking, and end-to-end metadata.
Discovery: Discovery is a general term for finding content which encompasses searching, browsing, content routing, and other techniques. Discussions of discovery frequently center on a consumer searching a public Web site. However, discovering content is much broader than that. The audience may consist of consumers, or it may consist of internal users such as researchers, designers, photo editors, licensing agents, etc. To assist discovery, PRISM provides properties to describe the topics, formats, genre, origin, and contexts of a resource. It also provides means for categorizing resources using multiple subject description taxonomies.
Rights Tracking: Magazines frequently contain material licensed from others. Photos from a stock photo agency are the most common type of licensed material, but articles, sidebars, and all other types of content may be licensed. Simply knowing if content was licensed for one- time use, requires royalty payments, or is wholly-owned by the publisher is a struggle. PRISM provides elements for basic tracking of such rights. A separate vocabulary defined in the PRISM specification supports description of places, times, and industries where content may or may not be used.
End-to-end metadata: Most published content already has metadata created for it. Unfortunately, when content moves between systems, the metadata is frequently discarded, only to be re-created later in the production process at considerable expense. PRISM aims to reduce this problem by providing a specification that can be used in multiple stages in the content production pipeline. An important feature of the PRISM specification is its use of other existing specifications. Rather than create an entirely new thing, the group decided to use existing specifications as much as possible, and only define new things where needed. For this reason, the PRISM specification uses XML, RDF, Dublin Core, and well as various ISO formats and vocabularies.