Multi-representation Database

Imagine producing several different types of maps using different style sheets - different symbology, different scales - from the same data model and in the same database. How does this work?

By using a true, multi-representation data base, it is possible to have one set of data for an entire city, country - even the world - and use the same data to create and produce many different maps. Changes made in one place are propagated to other representations in the same data base. If new data is introduced to the data base as an update, a history is kept for each piece of data that is introduced AND a relationship is built with the original piece of data. It is possible to see what has changed and whether or not the changes are significant. This is critical in the automatic generalization process since re-generalization can be carried through by analyzing the significance of the changes over time.