Toward design patterns for dynamic analytical data visualization
Chen Hong · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2004
This paper advocates the study and use of visualization design patterns to improve development productivity and usage effectiveness in dynamic, analytical data visualization. Nine visualization design patterns are presented formally using the current de facto pattern description language. Organized in three categories (data, structural, and behavioral), these patterns summarize many common practices and techniques used in the process of dynamic, analytical data visualization. A relationship diagram is also introduced to illustrate the common relationships and uses of the patterns. Driven by the study of design patterns, a simple, yet powerful, architecture design for a dynamic, analytical data visualization library is proposed. The fundamental characteristics of the design are component-based, data centric, and layout-aided.