One of the things we have noticed is that there are lots of interesting areas for companies to explore in terms of process and technology, but not a lot of resources that can be devoted to exploration without a cost justifiable outcome. So for example, someone might have a departmental budget of $10k or $20k to look into a new area, but that would not be enough to adequately explore the issue at hand. However if several organizations pooled those resources for a common research agenda, then each could have the benefit of a larger project for the cost of their contribution.
Based on conversations I have had with a number of firms, Topic Maps may be a good area for us to organize a syndicated research project around. Here is a rough description of the project. Contact me if you are interested in participating.
Topic Maps: Cost, Value, Capabilities
Topic Maps have been around for a while, but there is still a great deal of confusion in the marketplace as to their practical benefit and technologies to support them.
Different people define them in different ways. There are no clear examples of their widespread usage. Taxonomies are not a silver bullet. They are a means of organizing concepts but not an actual structure to access documents and artifacts. Indexes are hand crafted and specific to a body of content. Taxonomies are reusable, but can lack context. Indexes are not reusable but they contain inherent assumptions about the user and about the needs of the user relative to the content.
Topic Maps are thought of as something between taxonomies and indexes – we are able to retain the semantic relationships and context of a taxonomy or thesaurus, but also can point to specific instances of information.
Topic Maps are also used in search contexts. A topic can be created based on a custom set of rules and essentially becomes a stored query executed against a domain of information.
Though there are a number of organizations that claim to be using Topic Maps, what are these implementations really about? Are they useful, are they worthwhile? Are they cost effective? Is a Topic Map just a name for a bunch of custom code or is it truly a framework that is scalable and flexible? This project will seek to answer these and other questions by pooling resources of multiple participants and executing a research study that will examine:
- Definitions: what are the various definitions of topic maps?
- Technologies: what are the classes of technology that enable or leverage “topic maps” , what are some of the open source and proprietary players?
- Practical applications: what applications are most appropriate for Topic Maps?
- Usage: what organizations are using topic maps?
- Deployment: what are the practical implications of deployment?
Filed under: Semantic Web |