Two elements greatly assist in sharing environmental information: the use of controlled vocabularies for classifying content, and having environmental data and metadata be hosted in a distributed, local manner. As has long been recognized by librarians, using controlled vocabularies, thesauri, and web-based ontologies facilitates structured creation, search, and retrieval of information. Moreover, data owners and custodians often prefer maintaining close control over datasets and catalog information rather than diverting these to a centralized clearinghouse for outside access; developing mechanisms for distributed search and retrieval of data is important here.
Both of these elements are captured by the technologies emerging with the Semantic Web, and our informatics work follows in this vein. Standard vocabularies such as Dublin Core are now commonly used, and the Semantic Web offers schema languages such as OWL and SKOS which allow communities to define their own local ontologies and vocabularies. And protocols such as RSS, which is widely used for syndication, provide an excellent model to build upon for distributed retrieval of information.
Under an NSF-funded project called SPIRE, we have been working to develop schemas, thesauri, and other applications for the environmental Semantic Web.
CAIN Informatics Resources
