Flexible and Personal Data Exploration with Mashed-Up Cartography

Jason Dykes,  PhD

Developing technologies and standards are assisting with the generation, collection and documentation of masses of data. Consequently data and metadata through which it is georeferenced are increasingly prevalent and accessible, often through open online interfaces. This includes formal geographic information and the less formal "geographic Web" of volunteered georeferenced content. 
    Opportunities to explore the geography of the newly available data sets result, along with needs to develop techniques through which this exploration might be achieved. Methods that take advantage of human skills in visual processing can contribute. Visual approaches that offer fast responses to data dense graphics can enable analysts to: filter large data sets by time and space; relate diverse data sets according to geography; use geography to identify spatial trends and patterns — both within phenomena (that may relate to spatial trends, errors, anomalies, or issues of data integrity) and between phenomena (that may lead to ideas about cause and effect, or suggest that relationships vary geographically). 
   The complexity, novelty and unknown nature of the data sets being explored through visualization require highly flexible and adaptable approaches that enable analysts to use their knowledge, insights and skills to develop and modify maps and graphics as their queries, ideas and understanding develop. These temporary, exploratory maps are generated and modified through flexible combinations of technology and interface to guide the thought process and apply tacit (spatial) knowledge to spatial information. We may consider this egocentric, rather than automated, cartography. 
    A number of open technologies allow us to combine well documented data formats, sophisticated applications programming interfaces and interesting data sets for visualization. A selection of loosely coupled and loosely bounded applications that combine functionality and content from diverse sources for data exploration provide examples of how these technologies might be used in egocentric exploratory cartography. These “mashups” have been developed to generate insights from data sets that are traditional and formal, that are historic, that relate to mobile information use and that originate from user content volunteered and made available through “the geographic Web.” A number of novel layouts and interactions have been developed during the exploratory processes. These can be considered 'mashups' of ideas, knowledge and techniques from cartography, geography, information visualization, statistical graphics and elsewhere. They include tag maps and spatial tag clouds, data dials, maptrees, geowigs and dynamic links between abstract representations and realistic imagery.

   Characteristics of these examples of exploratory EgoCarto—personal and individual 'mashups' of technology, data and cartography—will be presented and critiqued.