parid2176 | Mon, 24 Feb 2003 12:55:12
Syntactic constructs (or other formal and addressable information objects) that appear in representations of topic maps. The existence of such an addressable object in a representation of a specific topic map is regarded as "demanding" the existence of a corresponding node in any topic map graph that is constructed on the basis of such a representation. The purpose of identifying the node demanders that can be found in instances of (for example) a topic map interchange syntax is to allow, effectively, the subjects of the nodes that they "demand" to be addressed in terms that will be honored by independent implementations. When a specification for a topic map interchange syntax includes a specification of its node demanders, topic map authors can create assertions whose role players are the topics specified by topic maps expressed in that syntax, with the expectation that implementation of that syntax specification will behave predictably.
An information object in a specific topic map that requires its representation as a node in a topic map graph.
The foregoing catches the gist of this paragraph, sans the explanation of the how "node demanders" compell compliance by other implemenations. Seems like a lot of overhead to say that if nodes in a specific topic map are not treated by an independent implementation as specified in the original topic map that the results will be different from that of the orginal topic map. Trivial to say that if an independent implementation does not honor the roles defined by a topic map, that the behavior will vary from the original topic map.