There are many models of hypertext, distinguished by a number of factors su
ch as the underlying semantic data model (link typing and node typing), the
degree of dynamic linking in the hypertext, and how dynamism and other beh
aviours are implemented. This essay examines a particular approach to dynam
ism in hypertext, based on the degree of similarity between a text passage
in a source node and the text of a target node. It reviews work carried out
over the past decade in creating systems for markup-based querying and dyn
amic hypertext, with particular emphasis on a model of dynamic hypertext th
at computes hypertext links on the fly using queries.