In a geolibrary, the main components supporting resource discovery are:
• Metadata bases structured after proprietary or standard format (and holding discovery metadata)
• A query model that can be very close to the metadata base structure
• Contextual databases like a thesaurus or a map
The expression of the user need relies on the query model. This is always a set of elements comprising the geographical area of interest, keywords or themes, and the data producer. Each element of the query model is associated to specification modalities, like the use of contextual databases to support the specification of the element in the user query. Typically, gazetteers and maps support the user specifying his geographical area of interest by writing down a place name or delineating a zone on a map. A thesaurus supports the specification of themes or keywords.
The obtained query is at last mapped into the formal query models of the metadata bases.
The most famous geolibrary is the Alexandria Digital Library (ADL, www.alex- andria.ucsb.edu). The ADL query model is called the Bucket Framework . The Bucket Framework specifies a set of buckets and the mapping between each bucket and corresponding elements in various standard metadata models. The buckets are: geographic locations, dates, types (e.g., map, aerial photograph), formats, assigned terms, subject-related text, originators, and identifiers.