ESSENTIAL CHARACTERISTICS OF KNOWLEDGE MAPPING TECHNOLOGIES

Given that knowledge mapping may involve analog and/or digital culture, it is important to identify the essential characteristics of a an effective knowledge mapping system for both. The discussion which follows attempts to establish a list of such characteristics which evolved from my work with Professor Robert Wright. It was concluded that there were at minimum, 4 essential characteristics. We converted these to constructs which could be used to measure or assess the viability of any given knowledge mapping tool or system.

• Sustainable - This construct measures to what degree the knowledge mapping system is self-organizing. Self-organizing systems are far more sustainable (i.e., anti-entropic) than systems requiring external monitoring and control. Ideally, the system would contain within itself its own evolving blueprint for growth and adaptation, much like DNA in living organisms. Functionally, this translates into a sophisticated means of tagging knowledge and information, coupled with a feedback loop to reward contributors.
• Open - This construct measures how explicit the system is to users. It also involves the depth of knowledge contained, since it is possible to have a highly explicit, but superficial representation of knowledge (much like this condensed paper). An ideal system would be completely open, within the limits of personal rights to privacy and confidentiality.
• Accessible - This construct measures the ways and means by which the knowledge mapping process may be engaged. An ideally accessible system should make for ease of connection, permit a natural language interface, involve no proprietary hardware, prove inexpensive (affordable) to any user, and provide unrestricted access to resident knowledge mapping processes.
• Robust - This construct measures the adaptability of the system in all of its aspects. The ideally robust system should enable any user, independent of background, bias or perspective to engage the system according to their own terms of reference. Technology should not encumber users with adequate entry level knowledge.

Clearly, essential characteristics for a system of this nature should also be embodied in the tools it engages. An ongoing collaborative research project by the New Media Group of the Rogers Communications Centre at Ryerson Polytechnic University, entitled nViews, is beginning to shed some light on these criteria.

The nViews Project

Within the knowledge mapping framework advanced in this paper, nViews may be viewed primarily as a construct tool. In this case, the constructs involve documentary processes within a multimedia environment. nViews users will likely continue to rely on dialogue, operational and synthetic tools developed outside of the nViews project to resolve problems, deal with issues and explore opportunities. This limited scope in no way diminishes the significance of the project, since documentary knowledge mapping represents a considerable advance over the present system of media production, androgogy and internship.

To date a number of developments in thinking about knowledge based tools have taken place:

1. the tagging of knowledge and information integral to the knowledge based tool is arbitrary, provided the user can apply their own individual tagging system (my implicit map of a library bears little relation to its indexing system);
2. the determination and representation of essential knowledge will remain the responsibility of each user, and this cannot be presumed by the system - the idea here is to create a prosthesis for human knowledge mapping, not artificial intelligence.
3. the first user will establish the default tagging system for the next user and so on, such that the default tagging system will represent an inherited mosaic of contextualizations (users will automatically gain part if not all of the recipes of previous and concurrent users);
4. representation of these various knowledge maps, either individual or hybrid, will be at the discretion of the user, again with a system of inheritance built into the process (any perspective is possible given a particular bias or cultural preference);
5. the degree of exposure (explicitness) remains at the discretion of the user - presumably less explicit knowledge maps will not gain prominence over fully open processes; and
6. the knowledge mapping tool only deals with the maps (DNA fingerprint of the knowledge mapping process), the actual information resides at its source or some external repository, significantly reducing storage requirements by adopting a distributed structure, as well controlling transactions involving remuneration;
7. the system should be designed not to preclude the adaptation of "ideal" technologies which may not yet have been developed, hence it must be open ended and flexible; and
8. the accounting system for the protection of intellectual property must be integral to any knowledge mapping tool and not pose any additional burden on knowledge workers.

These constructs, which were developed during the initial stages of the nViews project, are likely to evolve as nViews proceeds. To a large extent, this tool and its application will also evolve to reflect the particular focus of the enterprise, or enterprises, employing it in their documentary knowledge mapping processes. A general schematic for how these processes might relate to analog and digital culture and communications is depicted in Figure 3.

Many representations are conceivable and plausible. At this point, the only apparent certainty is that any knowledge mapping system will have to resolve the issue of intellectual property. In the information age, this issue cannot be avoided if information technologies are to become broadly based. For those types of expertise which reside primarily within a digital construct space, a tool like nViews should permit the explicit citation of intellectual property, and hence the basis for an accounting system for the apportioning of dividends or royalties. This signature must remain indelible, akin to a DNA fingerprint, and permit the full tracing of the genealogy of knowledge and information. But this is more of an issue for the electronic media industry, to be standardized, for instance like MPEG, and beyond the scope and resources of the nViews project.

The nViews project team felt, at the outset, that it was important to first recognize a broader context for this project, but then openly declare that nViews could not be all things to all people. It remains important, however, to keep speculating on future possibilities, no matter how impossible they may appear from this point in time.

From a more realistic perspective, nViews will likely remain a highly specialized multimedia construct tool, serving as a component in a more sophisticated, composite knowledge mapping system. There are still many problems to be overcome before nViews could be used in a mature knowledge mapping system, and these will continue to be explored throughout its development.

In the meantime, I will venture to speculate that the Web will likely become the primary dialogue tool for knowledge workers, whereas a large number of flavours of construct tools, such as nViews, may be expected to follow. With the insight gained from these two knowledge mapping spaces, there may likely be some cause for re-examining appropriate operational processes. The problem of synthesizing expertise and explicitly representing this top sight will always remain a formidable challenge.

This paper was intended to provide one view from a knowledge mapping explorer, simply as a departure point for further dialogue. nViews, or for that matter knowledge mapping, is not being portrayed here as the key to the universe of knowledge. But it is difficult to imagine developing appropriate tools and processes in the absence of any explicit knowledge mapping models. At some point, there will have to be some shared basis for adapting to the world of knowledge and information we are creating, if it is to be sustained.