Resources and metadata Two basic sorts of objects are considered in this chapter: resources and metadata. A resource can be an electronic document (e.g., a jpeg image, a software

component, a relational database, a text file, a Web resource, etc.), or a non-electronic document (e.g., a book, an argentic picture, etc.), or even any “object” (e.g., a historical building, a tourist resort, a drawing, etc.). Thus, a resource is a very general notion, and we do not restrict it to documents. On the other hand, we greatly restrict metadata. The metadata considered in this chapter are computer data structures or,

more precisely, mathematical objects that can be implemented as computational data. A metadata is always associated with a resource. Thus, we consider two sets. A set of resources, sometimes called primary data when electronic data are considered, and a set of metadata associated with these resources, sometimes called secondary data. Each resource is identified by an identifier, which is used in the metadata

base for referencing the resource. A resource can be decomposed, e.g., a book can be decomposed into chapters. But hereafter a part of a resource is considered as a resource, i.e., we do not necessarily take into account, in the

annotation base, the structure of a decomposed resource.