Policy Driven Spatial Data Update
David Skea-INFORMATION MANAGEMENT BRANCH, and Yao Cui, BASE MAPPING AND GEOMATIC SERVICES BRANCH MINISTRY OF SUSTAINABLE RESOURCE MANAGEMENT PROVINCE OF BRITISH COLUMBIA
Spatial data update can be viewed as a series of automated ETL (extract, transform and load) processes where each step in the series represents a move from one persisted state to another. For example: a series of GPS readings are processed into a lake edge observation, a set of lake edges are used to form a lake feature, a lake along with inflow and outflow positions are used to define a medial-axis skeleton, and lake skeletons along with stream edges are then used to build a stream network. GPS readings, lake edges, lakes, lake skeletons and stream networks are all objects that will be stored in a database. An ETL process can promote an object from one persisted state to the next only if the object meets a number of predefined criteria. These criteria and associated transformation processes (e.g., skeleton generation) makeup the update policies.
This paper reviews work done over the last year to implement such a policy driven update model for the management of British Columbia's base mapping data. The policies are stated using XACML, the transforms processes use the WPS specification and data transport and persistence is done using a WFS and GML