Iconographic visualizations of data employ icons or glyphs, graphical objects whose visual attributes are
bound to the data. Researchers have already explored a wide range of such data representations.
[10] used hundreds of highly detailed icons to show velocity, magnitude, and direction of flow
in a plastic injection molding process. [14] used thousand of glyphs whose color and shape
distortion represented stretching, compressive and shear forces in a display. In these applications the
information conveyed by individual icons was of paramount importance. [28, 17] used
several thousand icons to represent data, but here the aim was to produce collective effects such as
gradients or islands of contrasting textures which correspond to structures in the data. In their application
of icons, perception of the collective effects are more important than the appearance of individual icons.
Pickett, Grinstein and Levkowitz [28, 25] have shown that it is possible to get
meaningful displays of data structures by using icons to generate textural representations of the data. The
potential advantages of this approach to reveal high dimensional data structures has been repeatedly
demonstrated [29, 12, 25, 11]. Despite the high promise of this approach, its use has been limited. Recently, Pinkney
[30] has proposed a formal model of icons and icons interactions based on L-systems
[31] that provides for much greater flexibility in its application. This course of research will
apply Pinkney's technique to expand exploration with, and the ultimate utility of textural icons displays.