BRIEF REPORT

Scale Is Introduced In Spatial Datasets By Observation Processes



Scale Is Introduced In Spatial Datasets by Observation Processes

The article “Scale Is Introduced in Spatial Datasets by Observation Processes” by Andrew U. Frank was presented in 6th International Symposium on Spatial Data Quality. The paper explores the process of geographic data collection and show how scale is introduced during the observation process, and should be carried forward as a quality indication. An analysis of the properties of real (physical) observation processes reveals that physical observation processes introduce a scale into the observation (Frank, 2008).

Th paper starts with a brief review of tiered ontology, which is necessary for the analysis. Section 3 lists the information processes that link the tiers and gives the framework used. Section 4 discusses briefly accuracy and shows how random imperfections in the observations influence the formation of objects and the values for their attributes. Section 5 looks at scale, produced by the spatial and temporal extent of the observation as a second source of imperfection in the data. Convolution gives a formal model for this effect. The influences of scale are so defined in the process from observation to object related data.

The paper also discusses about the digital data that the digital geographic data comes in different qualities, and applications have different requirements for the quality of their inputs. A common misconception is that better quality is always preferable, forgetting that better quality means more detail and therefore more data, longer data transfer and processing time, etc. Traditionally map scale was used to describe the quality of geographic data comprehensively. With the reduction in scale, expressed as representative fraction, comes automatically a reduction in detail, described as cartographic generalization. The beginning of such a theory is attempted here.

The present contribution extends Goodchild's observation and states that ...