Modular Techniques in Information Visualization

The use of visualization to explore and understand data is often partitioned into two areas, scientific visualization in which the data sets are typically derived from measurements or simulations grounded in physical space, and information visualization where data sets are defined over abstract spaces. Although there are certain pragmatic differences based on the way that visualization is used, this paper argues that there is interesting progress to be made by ignoring such distinctions, and working with a general model in which visualization is about representing structures within particular kinds of space. This applies both at the conceptual level, and at the level of practice and implementation. This paper sets out the motivation for thinking in these terms, and describes initial work on using a toolkit, designed primarily for scientific and engineering applications, in a rather more abstract domain, graph visualization.