If we adopt the traditional notion that rhetoric is argument, then we need to find ways to teach task-oriented, goal-driven reading that co-opts the "clickable classroom" into a site for disciplined inquiry. But if we accept that this emphasis on discipline may be no more than a papyrocentric attitude left over from a print-based culture, then we may want to substitute a more exploratory view of rhetoric and hence of reading.
Pursued to its logical conclusion, an exploratory view of reading would entail a significant change in the ways in which reading is generally taught. A classic strategy, for instance, is to require a precis or summary of material. The most important aspect of a summary-writing exercise is to isolate the conclusion, main claim, or essence of the work, and then to use that information to separate main arguments from secondary ones. Behind this lies the tacit assumption that there is a point-to-able difference between claim and support, between main arguments and secondary ones.
Constructivist views of reading have already rendered these notions suspect. More radical forms of hypertext demolish them entirely. Hypertexts not only allow different paths toward their conclusions: they may have several conclusions, or no conclusions. They encourage juxtapositions of various points of view, circles of comments and metacomments, endless deconstruction of their own propositions. The author and the text, always suspect entities, are revealed as so much less important than the reader that attempts to summarize them become exercises in frustration and futility.
This does not mean that readers should not compare readings. In fact, it makes these activities more important than ever. But it means that comparison of readings should be comparisons of constructed meanings, of alternate paths and alternate conclusions, a celebration of difference rather than convergence of meanings.
If this seems altogether too chaotic, maybe it means that hypertext just isn't a good medium for argument in the first place.