One of the results of this experiment is that the students were able to develop well-organized, readable documents without either a complete specification of the documents' content or of any particular knowledge of hypertext organizational principles. Given the ease of learning HTML, this suggests that students in any class where information could benefit from similar analysis and organization - literature, history and the social sciences are obvious examples - could be reasonably assigned non-trivial hypertext projects for a semester course.
However, outside of the novelty of such a project, one should also be concerned with its effectiveness: How well did the students learn the new material in comparison to previous students who did more traditional assignments? The results from this experiment are unequivocal in this respect. The course professor described the experiment as a ``great success,'' pointing out that there was a measurable improvement in the individual papers submitted by this class compared to previous ones. Thus, the hypertext projects increased not only the students familiarity with the material but improved their analytic abilities as well.
One of the reasons for this increased understanding is based on the organization of the information itself, and not simply the reinforcement of material that building the documents obviously required. In this case, the students were not only developing a way of organizing information for a course but actually developing a way to analyze cognitive architectures which, heretofore, had been largely ad hoc, based on the designers observations about what had worked (and what had failed) in the design of individual architectures. Thus, the organization of these hypertext documents may provide a methodology for the evaluation of future cognitive architectures.