Rethinking document structure in a global hypertext environment: toward a model of nested conceptual units of information on the World Wide Web

The World Wide Web represents a social, economic and technological milieu for the public availability of information. This information is available in what can be thought of as Web information units. An information unit, considered as an abstract structural concept, consists of information within given physical and conceptual boundaries. Information units are frequently nested into larger units for purposes such as distribution to and retrieval by readers, archiving, republishing, and so forth. In an established and relatively stable system for the public availability of information units, such as the print publication system, the boundaries of information units have become conventionalized and identified with specific vocabulary, like book, journal article, report, document, and so forth. In the system of public availability of information now developing on the World Wide Web, the conventional understandings of boundaries of information units are undergoing redefinition in the social, economic and technological conditions of the emerging system. There is a need for concepts and terminology representing the structure and relationships of publicly available conceptual information units on the Web. The paper proposes a four part structural model of nested information units on the Web: pagefile, document, package and collection. It concludes with a series of implications of this model, with respect to page self-identification, bibliographic control of Web units, and Web unit citation.