In the previous sections we saw that generic link following succeeds by matching representations of user selections with the source anchors of links. In the case of text, an exact match or word stem match is appropriate whereas for non-text it is necessary to use similarity matches. However, even in the case of text, the basic generic link has a severe limitation. For the generic link to be successful the user must choose the same word or words as those used when the link was authored. Synonyms and semantically linked concepts are not catered for in the current implementation. If a generic link is authored on the word car, it can not be followed from the word automobile. For non-text media there are equivalent problems. For example, for images, if a link is authored on the shape of an object from one particular view it will not be possible to follow it from an image in which the view is substantially different.
One way to approach a solution to this problem for text is to introduce a digital thesaurus. This should contain all the vocabulary (words or terms) appropriate to the application domain, arranged as a network with at least the following relations: broader term, narrower term, equivalent term and related term. The thesaurus could be invoked either at the link authoring or the link following stage or both. In the first case, if the thesaurus is invoked when a generic link is authored, links could be automatically generated and stored from all terms equivalent to the selected source anchor term. This would increase the linkbase size but have the advantage of increasing the speed of link following. Alternatively, the authoring stage could remain unchanged, ie a single link is generated from the author's selection, but at the link following stage the user's selection is expanded via the thesaurus to a set of equivalent terms and each is compared with the linkbase source anchors to find any available links. This will be less expensive in linkbase storage but will slow down the linkbase search.
More elegantly the thesaurus could contain an indicator of a preferred term for each set of equivalent terms. At the authoring stage, the source anchor term is switched to its preferred term via the thesaurus and at link following, the user's chosen term is also switched to its preferred term, via the thesaurus, before the linkbase search is commenced.
All transactions via the thesaurus described so far can be invisible to both author and user, providing an appropriate thesaurus is available containing all terms likely to be encountered. Indeed, within the Microcosm architecture, the thesaurus could be introduced as a filter between the viewers and the linkbase, without substantial disturbance to the current architecture. However, an additional level of functionality may be achieved if, at the link following stage, the user is given access to the thesaurus to broaden or narrow down the generic links from a particular selection. For example, a user reading a document on analytical chemistry may wish to try and follow links on mass spectrometry. If an option to view the thesaurus at this point is available the user may choose narrower terms such as high resolution mass spectrometry or broader terms such as spectrometry to act as the selection for link following. Indeed the user could also choose to switch on a particular level of automatic broader or narrower term fan out for selections when generic link following. In this case, whenever a term is chosen, all broader terms or all narrower terms or both could also be compared with the linkbase source anchors to find appropriate links to follow.