The topic of spatial cognition (and closely related, naïve geography) was relevant to the issues discussed by both Elwood as well as MacEachren & Kraak. The ways humans learn geographic concepts and reason about space is required for geovisualization to “handle qualitative forms of spatial knowledge” (Elwood, 259) and for building “human-centered approach to geovisualization” (MacEachren & Kraak). I believe developments in this field are urgently needed and have far-reaching implications not only for geovisualization but also for building ontologies. In fact, Smith and Mark also touch on the lack of research by stating “We know of no data on the ages at which young children acquire or master the basic concepts of naïve geography and the associated kinds of objects…” (10).
With a growing amount of geo-located SMS, pictures and videos, how can we process these qualitative information without grasping how it is that the contributors comprehend their surroundings? Since users are also contributors in the Web 2.0 environment, it is evitable that we must dedicate resources to understand these users. For instance, how do people learn and remember directions? How do people from different cultures use landmarks, whether natural or man-made? Only by understanding how people build their relationship with geographic space can we take more initiative in the geovisualizing process and derive meaning out of spatial descriptions (near, far..). As a side note, I imagine it would also be important to first identify what the source data was initially intended for because the context could influence how spatial forms are perceived and described. For example an emergency text message and a text message trying to rent out an apartment could be very different — the first message is influenced by panic and thus, the users might have a distorted conception of distances whereas the second message is motivated by the intention to sale and thus everything might be described as “near” the apartment.
-Ally_Nash