Intelligent Diagrammatic Environments
Bernd Meyer1, Hubert Zweckstetter1, Luis Mandel2, Zoltan Gassmann1
1 Univ. of Munich, Dept. of Computer Science, Oettingenstr. 67, 80538 Munich
2FAST e.V., Arabellastr. 17, 81925 Munich, Germany
contact email: email@example.com
Graphical user interfaces have become an integral part of almost every modern application type and it can be claimed that they are among the driving forces that have made the computer accessible to non-expert users. However, comparing the use of graphics in existent user interfaces with that in non-computer-based work, the inadequacy of standard GUIs for complex visual communication is revealed: Most GUIs are still WIMP interfaces centered around such simple interaction devices like icons, buttons, menus or image maps. On the contrary, in non-computerbased work rich and highly structured graphical notations prevail. There are diagrammatic languages in almost every technical discipline, for example circuit diagrams, architectural floor plans or chemical formulas, and modern software engineering is embracing all kinds of diagrammatic specification methods. Likewise, non-technical fields use their own well-established diagrammatic systems, for example choreography notation. But even in such application domains where diagrammatic notations are a natural element of discourse and where their meaning is well-understood, they are still rarely utilized to their full extent in the humancomputer interface. It is hardly ever possible for the user to communicate with a computer just by sketching an annotated diagram like two human experts would do in their discourse. This poor integration of richly structured visual communication into graphical user interfaces is mainly due to a lack of method and tool support for the implementation of the required interfaces. While countless specification mechanisms and interface builders for WIMP interfaces exist, there are almost no tools and only few specification formalisms for building intelligent graphical interfaces for such richly structured diagrammatic notations.
This work is supported by DFG Grant Wi-841 and Bayerische Forschungsstiftung.