User-Centered Approach on Geodata

Audio from every talk at last week’s IDEA2006 conference (conference on designing complex information spaces of all kinds) are now available online.

Urban Mapping‘s Ian White gave a talk on the design of data (MP3).

Ian White: The Design of Data (PDF)

In short, Ian talks about mapping complex georeferenced data and how the design of data influences the experiences the users will have. Prior to any product definition he categorizes data in their context of use:

• Mode (driving, public transit, bicyle)
• Environment (urban, rural, indoor)
• Domain opportunity (mixing mode and environmeent, e.g. public transit urban)
• Constraint (tempoal, technical logisical)

From his experience with print (i.e. Panamap) and his “quest to sell polygones” (i.e. a geospatial database of neighborhood boundaries they license) Ian mentioned a couple of issues that are right on the spot with what I do:

‘How Location Aware are You?’

Geo-aware device often offer a “range of uncertainty” and we know that we do not know how certain you are (knowledge typologies). It would be nice if you could make it explicit.

Centroid issue

In practice, a neighbor is defined with average centroid based on population density and then a radial curve is drawn. This barely represents reality in many cases and in the context of use many time useless.

In the field of building application taking advantage of geodata, Ian advises to

• Study the context of use of the users (qualitative data)
• then segment it rigorously
• Design is a process
• iterate!

It is really very similar to my design-science research approach.

Relation to my thesis: I am hardly interested in how the design of data influences the experiences the users will have. I am not sure I will apply methods that can answer that. Context of use is something I plan to categorize as well. The temporal (4th) dimension of data was barely mentioned, while I see it as one of the 3 main source of uncertainty. The problematic of defining neighborhood and how it should match the user’s perception of this space provides me of an example of uncertainty that comes from the processing of the geodata.