Uncertainty and geographic information: computational and critical convergence

The relationship between critical geography and geoinformatics has recently entered a new phase of cordiality and constructive discourse, exemplified by the NCGIA Initiative 19 [27], the Varenius project [21] and a variety of journal special issues on the topic of GIS and society (eg [28]). This new-found entente has opened the way to the re-evaluation of existing GIS-based research and to much closer cooperation between the two disciplines for the exploration of new areas of common interest. Within this context, this chapter represents an attempt to fuse of certain aspects of critical thinking on GIS with the development of new computational approaches to uncertainty in geographic information. Underlying this fusion is the contention that both subjects stand to gain from closer cooperation [41]. Such a fusion may provide a route to a more complete style of GIS-based research, acknowledging some of the important societal issues connected with GIS at the same time as accommodating some of the practical restrictions connected with computational systems. Amongst the variety of different critiques of GIS surveyed in [42], is the tendency of GIS to promote certain types of representations, or “ways of knowing” [39], over others, for example in the association of GIS with positivist modes of thought [32]. In an attempt to combat such tendencies, John Pickles ([39]) calls for a more collaborative “pluralistic GIS”, representing multiple perspectives and containing conflicting information. Bondi and Domosh [6] highlight the contradictions that arise when GIS ignores the pluralistic and treats geographic information as objective, uncontroversially ‘true’, and free of any context. Similarly, Haraway [26] explores the importance of “partial perspectives” in understanding information. This paper argues that there is some consonance between such critical geography research and current geoinformatics research into uncertainty in geographic information.