When one considers the limitations of the paper medium in communicating knowledge, and contrasts this with the capabilities of current multimedia computer and communication technology, the incentives to make major changes in scholarly publishing are very high. Books and journals primarily support typographic text and monochrome diagrams and photographs, and require separate indexing for purposes of contextual search. Computer-readable multimedia documents support the same quality of typographic text, color diagrams and photographs, sound and video, animation, simulation, computation, and are intrinsically contextually searchable. They can be produced and distributed on CD-ROM at a substantially lower costs than paper books or journals, and can be read on personal computers and selectively printed on laser printers. They can also be produced and distributed on the Internet, providing an active medium supporting many forms of scholarly discourse.
The objective of the KSI research program reported in this article is to provide a next generation of open architecture knowledge support systems (Gaines, 1994a) supporting collaboration in scholarly communities through information technology with a view to achieving systematic acceleration of human knowledge processes. Specifically, a new layer of knowledge level tools has been developed that complements existing list servers and electronic document archives with research coordination facilities. The tools are designed to support the knowledge processes of scientific communities through active document, hypermedia, concept mapping, knowledge representation and active agent technologies. They use existing Internet protocols for mail, file transfer and hypermedia to integrate with related systems. They provide import and export facilities for common document formats to integrate with existing document production. The data formats are designed for parallel use on CD-ROM to provide alternative means of distribution to the Internet.
The pragmatic objective is to achieve widespread use of greatly improved facilities for scientific discourse and collaborative project management. A deeper objective is to study the impact of the proposed technologies on actual research practice with a view to designing a next generation of tools targeted on the changes in patterns of research behavior already apparent as new information technologies have come to play a significant role in scholarship (Manitoba, 1993; Renear and Bilder, 1993) .
Figure 1.1 shows the general situation and support technologies studied in this article. A goal-directed, geographically dispersed community is engaged in a purposeful activity involving communicative action (Habermas, 1981) --that is, discourse in which mutual understanding is a joint objective. The process of discourse in a collaborative community can be supported through technologies such as electronic mail (email) and mailing list servers. The product of this discourse is captured in multimedia materials and document archives whose dissemination can be supported through technologies such as CD-ROMs and the Internet file transfer protocol (FTP). Hybrid client-server technologies such as those of the World-Wide Web may be used to support both process and product through structured discourse with archives indexed and linked to a record of the discourse.
Figure 1.1 Supporting the knowledge processes of a dispersed community
The principles and applications of the knowledge-level tools developed by the KSI have been reported in a number of articles which are available in publications (Gaines and Shaw, 1993, 1994b, 1994a; Kremer and Gaines, 1994; Shaw and Gaines, 1994) and through the web. This article focuses on practice and experience with commonly available technologies and tools. Its objective is to help others produce multimedia presentations in QuickTime, on CD-ROM, and through World-Wide Web.