The creation of models of nature is the main objective of natural sciences. Without abstraction, however, models would be as complicated as reality itself; they would mimic nature without helping us to understand it. Identifying the essential features of the phenomena being described is therefore a crucial element of model construction. Unfortunately, an emphasis on objective, measurable characteristics, as promoted by current scientific practices, may lead in the wrong direction. An easily measurable characteristic may turn out to be irrelevant; on the other hand, a feature that eludes precise definition or measurement may be of central importance. The paper illustrates this thesis by referring to the modeling of natural forms and patterns (in particular, plants) using the formalism of Lindenmayer systems combined with computer graphics visualizations. In this domain both precise botanical data and artistic observations play an important role. This synergy gives a new perspective to the centuries-old question of the relationship between science and art in describing the world around us.
Przemyslaw Prusinkiewicz. In search of the right abstraction: The synergy between art, science, and information technology in the modeling of natural phenomena. In: C. Sommerer and L. Mignonneau (Eds.): Art @ Science. Springer, Wien, 1998, pp. 60-68.Download Postscript (1.0 Mb compressed, 3.8 Mb uncompressed)