The Components of Hypermedia

Figure 1.1 is a general model of a typical advanced hypermedia system, divided into its component parts. This model might be applied to describe a system such as IRIS Intermedia (Yankelovich, Haan, Meyrowitz, Drucker 88). The different components consist of nodes, links, hyperdocuments, diagrams, a search engine and a programming interface. The information itself in hypermedia consists of a number of hyperdocuments, shown in dotted areas in the figure. The hyperdocuments again are built by a collection of nodes and links. Each hyperdocument comprises an independent, limited topic, and each node is an independent information unit. Links handle a natural switch from one node to another, thus structuring the hyperdocument.

A user's access to information primarily occurs directly via the hyperdocument nodes, by presenting each node in a screen window. This is suggested in Figure 1 by that part of the user interface box which goes directly to the hyperbase. The user will then follow the links from node to node based on the information contained in each node.

For a number of reasons it is not always sufficient to follow links to access the desired information. For example, the user may have "lost his way" and is unable to locate what s/he wants in the hyperbase, or the structure in the document may be too difficult to understand. Therefore two alternatives exist for accessing information: using a database diagram, or simply diagram, and using a search engine.

A database diagram is a graphic overview of the hyperdocument. It allows the user to navigate directly between nodes in the hyperdocument, without following the links. Furthermore, such diagrams may help users navigate by allowing them to more easily understand the structure of the document. They thus simplify both navigation and orientation in the hypermedia database.

A search engine is a mechanism which allows a user to search directly for information in the hypermedia database. Search engines may find nodes of a certain type, or names or nodes containing specific information. Search engines therefore will normally disregard the existing hyperdocument structure.

Advanced hypermedia systems may have search engines which allow the formulation of direct queries in the hyperbase. Users may then describe precisely the information they want using a formal language. Using a structured query language may also be applied to filtering. The information found by the search engine may then be merged with the structure information of the hyperdocument, allowing the user to see only an excerpt of the hyperdocument. This permits the user to navigate in the usual way, via links and diagrams, but only accessing those parts of the hyperdocument which satisfy the formulated query.

Some hypermedia systems have special programming interfaces enabling the creation of special applications for the existing hypermedia system. Furthermore, the programming interface may be used to add new features to the system. It may for example be used to connect the hypermedia database to more advanced search routines, or to link it to other applications allowing these to access the information in the hypermedia database.