Stencil: A Conceptual Model for Representation and Interaction

Existing Information Visualization models provide insufficient support to visualization programmers in creating applications. They either broad and taxonomy based, or narrowly focused on isolated aspects of the visualization problem. Taxonomy based tools are good for categorizing what has been or needs to be done, but provide little help in accomplishing those goals. Narrowly focused models allow particular aspects of the visualization problem to be efficiently solved, but leave heavy burdens for integrating many such narrow tools together to solve the over arching visualization problem. This insufficiency of visualization models is most evident where interaction is concerned: It is often left as an afterthought. In this paper, we describe the Stencil visualization model, an intermediate model that covers many visualization specific issues. We argue that the Stencil model can guide visualization program construction through several stages of common application pipelines; thereby improving the resulting visualization products and reducing significant barriers to visualization adoption.