The Case for Grounding Databases

What is the intended interpretation of a geospatial database in terms of reproducible experiences? How should places on a digital globe be interpreted on the earth surface? And how can their spatial relations be reconstructed? How should road network databases be interpreted in terms of observable traffic infrastructure? And how is data about waterways and their depths to be interpreted in terms of observable water bodies? In this paper, I argue that successful information retrieval and querying of data in context-free environments requires that such data interpretations need to be effectively coordinated. I give four arguments why the current approaches to semantic engineering fail as methods in that respect, and why a ‘grounding’ approach to describe their semantics is necessary.

• Type: Conference
• Year: 2009
• Where: GEOS
• Authors: Simon Scheider
• Keywords: GEOS 2009, Information Management, Observable Traffic Infrastructure, Observable Water Bodies, Semantic Engineering Fail