Contents in Current Page:
Nodes are the primitive fundamental unit for organizing information and they are together with links, the two primitive constructs (according to the basic hypermedia model) of all Hypermedia systems so far. We can say that nodes act as a repository for pieces of information which are directly interelated. For example in a hypermedia system about the life of Aristotle, a node can include a text description, a sound information, a graphics picture and a digital video all about the Aristotle's birthplace, Stagyra. Nodes in basic hypermedia model serve as primitive untyped abstractions in the information space. This simple model has been proved to be inadequate in many situations. There are many issues and also questions relating to this inadequacy and possible enhancements to the node concept in future hypermedia systems.
Basic hypermedia model lacks of a composition mechanism for structuring, representing and dealing with groups of nodes as a unique entities separate from their components (Botafogo et all, 1991), (Parunak, 1991). Current hypermedia technology is not capable to represent, at least in a natural way, whole/part relationships. The requirement for a composition mechanism becomes critical especially if we assume that hypermedia systems will have to deal with a vast amount of information. Many questions arise when designing a composition mechanism. Can a given node be included in more than one composite ? Do links necessarily refer to a node per se, or they can refer to a node as it exists within a current context of a composite? Should composites be implemented using specialised nodes and links, or is an entirely new mechanism?
Another important issue related to nodes is the ability to have typed rather than untyped nodes. Most of current hypermedia systems have untyped nodes. They can only be characterised by the primitive information or combinations of primitive information (e.g. text, text and graphics) they include, or as annotation, reference nodes etc. Hypermedia authors cannot define node types beyond the types that are "compiled" to the system. For example, they cannot have for example a 'Painting' node type in a Hypermedia system for Impressionism.
Nodes are passive repositories of information. There is no mechanism to adapt the content and structure of a node to current context. Adaptive nodes might be significant in some task domains, (e.g instructional hypermedia). For example a node can include two different texts explaining nuclear power. The two texts can differ in the prerequired background from end-users. Thus, the same node can present different contents in a novice user than in a more experienced user about nuclear power. In the same way structure in which a node presents its information to users can be adaptive to current context.
There are two different types of user interfaces, related to the way in which a node is presenting in a hypermedia system. We have frame-based and window-based user interfaces. Nodes in frame-based interfaces take up a specific amount of space on the computer screen, no matter how much information they contain. Very often the size of the frame is defined as the size of the screen. Since the frames have specific and limited space, hypermedia authors have to split the information among different nodes. In contrast, window-based systems require users to use a scrolling mechanism to get the part of the node they are interest in. As we understand the choice of user interface style can have serious impact in node design and in quantity of information a node can include.