Search
Browse By Day
Browse By Person
Browse By Room
Browse By Research Area
Browse By Session Type
Search Tips
Meeting Home Page
Personal Schedule
Sign In
Social media offers a chance to collect data in “data scarce” places, but it also presents the temptation of a universal understanding of the use pracitces that produce these data. I explore the creation of a geodataset in Nairobi, Kenya collected through a crowdsourcing project in collaboration with a local communication platform, Ma3Route. Because almost all reports (which came via Twitter, Facebook and a mobile application) lacked geo-coordinates despite the capability to automatically include them, I devised a reverse geo-coding system that used landmarks and road names to triangulate approximate locations for the reported events (traffic accidents). Thus, reverse geo-coding sacrificed the “accuracy” of automatic geo-coordinates for the accuracy of representing places as they are described locally. Reasons for not including geo-coordinates (according to Ma3Route employees and users) could be concern about making one’s location public or traceable, concern about spending mobile data credit, or the result of a feature phone that does not have a GPS. Due to the prevalence of not including geo-coordinates and the range of economic, political and social reasons, a geodataset of only automatically geo-coded reports would be highly exclusionary. This paper explores how the lens of non-use implies an intentionality that empowers online behaviors that do not conform to expected, “normal” use (particularly from the Western/Global North orientation). Reverse geo-coding is put forth as a means of privileging local knowledge and practice over global norms. As such it is presented as an example of post-colonial computing (Irani et al 2010) in practice.