ow, while the twenty-first century is still young and hopeful, it's time to guess about trends in the geospatial realm. Rather than predicting the next killer app or showcasing a rising geospatial startup, I'd like to highlight an existing but neglected idea that I predict will become a mainstream favorite on the geospatial hit parade in the years ahead. It's an old idea that's finally ascending the ranks: geospatial data modeling.
Data modeling, with and without computers, is a long-established discipline in the sciences, the military, and the business world that duplicates realworld conditions with intelligent data and embedded rules about relationships between objects. Often, data models seek to predict the future by testing many different possible actions against a virtual reality.
Geospatial data modeling with commercial databases is a new step for our industry resulting from several converging trends: an increasing amount of good quality geospatial data, better access to that data by the public on an ever-faster vector-streaming Internet, the expansion of wireless technology, and the reliable interoperability of GIS packages with commercial databases. Also, GIS users have enough experience managing and analyzing their spatial data to recognize the benefits of storing not just data, but also data rules and relationships, in one package. Together, these changes in the technology environment favor the migration from standard filebased spatial data storage to intelligent spatial data models. Vendors and users alike are ready for this migration, so the shift should happen with increasing regularity in the years to come.
