Interest in Workshop Participation

and

Contribution to the Workshop

* To foster work on abstract, formal descriptions of the content of maps or geospatial data sets that reflect human cognitive patterns

* To explore the potential for isolating, recording, and using abstract rules in generating and exploiting digital geospatial data

* To develop a foundation for allocating features to one thematic coverage as opposed to another

* To identify research underway that DMA or its customers might be able to use

* To identify opportunities for DMA to work more closely with the research community to the benefit of researchers and practitioners

The move to multi-use collections of digital feature data has led to an effort to define the contents of data sets in terms of standard feature and attribute definitions, combined with enumeration of features and attributes and Boolean conditions for their inclusion. Such an approach misses the concept that a digital data set is more often a model of some geophysical or geopolitical reality; while it comprises features, it is more than a collection of features.

We are exploring the potential for using sets of abstract rules to describe the content of geospatial data sets. DMA could use such rules in several ways. One use would be to organize our data to make it as intuitive as possible for users to locate data they need and for them to understand what they need to know about the data to use it responsibly. In addition, they could be useful in improving the presentation of geospatial information. We would also expect to use such rules to organize feature extraction and maintenance. Finally, we would expect such rules to result in more elegant and powerful software for use with geospatial data.

The attached position statement describes a small ad hoc effort to explore the possibility of using formal grammars to describe data set content. We are interested in reactions to this effort and in alternative descriptions of comparable or greater descriptive power. Analysis of "naive geography" that affords insights into how people organize geospatial information, as well as how they manipulate it, has the potential to help surface underlying rules and to reveal what might be universal and what might be culture or context-specific.

I have done some preliminary sketches of how a grammar for geospatial content might work and will bring them to the workshop. I am currently working on strategies for populating and maintaining large, distributed, geospatial data repositories. As a producer of data sets representing varied thematic coverages, at multiple levels of detail, over diverse terrains, DMA offers an environment rich in examples and potential counter-examples to hypotheses set forth in the discussions at the workshop.