We present a taxonomy for Information Visualization (IV) that characterizes it in terms of data, task, skill and context, as well as a number of dimensions that relate to the input and output hardware, the software tools, as well as user interactions and human perceptual abilities. We illustrate the utility of the taxonomy by focusing particularly on the information retrieval task and the importance of taking into account human perceptual capabilities and limitations. Although the relevance of Psychology to IV is often recognised, we have seen relatively little translation of psychological results and theory to practical IV applications.
This paper targets the better development of information visualizations through the introduction of a framework delineating the major factors in interface development. We believe that higher quality visualizations will result from structured developments that take into account these considerations and that the framework will also serve to assist the development of effective evaluation and assessment processes.