Terra Cognita 2006

The common threads of information running through diverse data sets and domains are people and identity, money, time and place. While few applications areas are purely representations of geography alone, the concepts of location and place provide a supporting role to many different applications, and as such they are crucial to information integration. It is this concept of integration of information, whether it is represented on web pages or stored in databases, which lies at the core of the semantic web. Estimates suggest that up to up to 80% of all applications have a geographical component – for example, disaster management, transportation, the environment, location-based services, navigation, local search, insurance, retail, marketing, defence and security, asset management, planning and construction. In all these cases, the geospatial element is not the dominant factor; but it provides an important cross-domain framework to which the primary elements in each domain can be referenced.

The geospatial and semantic communities have each found their own way to the Web but continue to have difficulties in finding each other; so far we are the poorer for it. The Geospatial Web has raised the Service-oriented Architecture to a high art, but relies too much on a brittle mixture of syntax and private knowledge to communicate meaning with geodata. The Semantic Web has brought concepts and relationships to data, but struggles with both the realities of managing distributed knowledge and the inevitably complex spatiotemporal properties of most information. A recent six-month experience in conversation and collaboration between these two communities (under the auspices of the OGC–Open Geospatial Consortium) has demonstrated the value of getting members of these communities together in person to work at bridging the divide in concepts and priorities between them. However both the these communities tend to be dominated by the information systems aspects and lack the experience of those whose expertise lies in the actual problem domains that geospatial and semantic technogies are intended to address. Hence a third community needs to be involved – the domain experts.

This workshop will seek build on this valuable experience by bringing workers in all three communities together, not only to describe their own work but to learn more about the value of each other’s work. Exploration of the semantics of geographic and other spatial concepts can provide the Semantic Web community with fresh insights and inform the development of other domains which confront similar modelling and reasoning issues. Conversely, formal representations of geographic knowledge and fully machine-processable service models can expand the horizons of the Geospatial Web community and the domain experts as to the potential reach of geographic information. The goal of the workshop will in fact be to develop a joint strategy and roadmap for working towards terra cognita – a Geospatial Semantic Web or more precisely a geospatially enabled Semantic Web which incorporates the best contributions of each group. A bonus for workshop participants will be the opportunity to help develop the charter for a new W3C activity focused on geospatial aspects of Web standards. More about this will be reported as workshop preparations proceed.