Data and Task Characteristics in Design of Spatio-Temporal Data Visualization Tools

It is widely recognized that data visualization may be a powerful instrument for exploratory analysis. In order to fulfill this claim, visualization software must be carefully designed taking into account two principal aspects: characteristics of the data to be visualized and the exploratory tasks to be supported. The tasks that may potentially arise in data exploration are, in their turn, dependent on the data. Developers of visualization software need comprehensive and operational typologies of data and tasks, where the task typology has an explicit connection to possible data components and characteristics. In the paper we present three examples of task-driven design of visualization software for different types of spatio-temporal data. We demonstrate that, first, different exploratory tasks may be anticipated in these three cases and, second, different sets of tools are required to properly support exploration of the data, although there are some common techniques and interface elements used in all applications. Prior to the description of the examples we briefly survey relevant typologies of data and tasks available in the literature.