The goal of this NCGIA-Varenius workshop and research initiative was to find new automated methods for filtering large amounts of raw geographic data into more user-consumable forms of knowledge. This includes: i) spatial data mining; ii) content-based and knowledge-based retrieval; iii) development of multi-media spatial data types; iv) on-line analytic processing; v) refinement of non-parametric statistics; vi) incorporation of computational intelligence techniques (such as neural networks and AI expert systems) into spatial data analysis. A key objective of the "geographic knowledge discovery" (GKD) workshop was to bring together the diverse stakeholders in GKD and spatial data mining. This report provides an overview of the workshop, lists the workshop participants, and provides the position papers submitted in response to the open Call for Participation. The final section summarizes workshop presentations and discussion.