Many of the pieces that motivate graphics algebra have been lying around for decades: experimental design algebras, relational algebras, table algebras. These algebras emerged from separate disciplines, so that in most instances, researchers have been unaware of developments in the other disciplines. What is new about chart algebra is the explicit and formal equivalence between the data structures needed for statistical models and the methods for displaying them in charts. In a general sense, this equivalence allows us to think about manipulating data by manipulating statistical representation elements of a chart.
The GOG project has had several purposes. One, of course, is to develop statistical graphics systems that are exceptionally powerful and flexible. Another is to understand the steps we all use when we generate charts and graphs. This understanding leads to a formalization of the problem that helps to integrate the miscellaneous set of techniques that have comprised the field of statistical graphics over several centuries. Another purpose is to develop, ultimately, intelligent systems that can 1) graph data without human specification and 2) read already published statistical graphics to recover data and interpret relationships. Finally, we hope to define problem specification and user interaction in a way that enables graphics systems to be understood by ordinary people as well as by statisticians. By formally relating display structure and statistical models, we can produce environments in which users interact with data, receive guidance, and draw conclusions that are based on appropriate statistical inferences.