Automatic construction of user interfaces from constraint multiset grammars

Describes tools which automatically generate a sophisticated user interface from a constraint multiset grammar specification of a visual language. The user interface allows the user to construct diagrams in the visual language from primitive tokens such as text, lines, rectangles or circles. These tokens are incrementally parsed into sub-diagrams. During parsing, automatic error correction removes geometric errors, providing feedback about what has been recognized. The user may add or delete tokens and may manipulate components of the diagram. During manipulation, the user interface takes account of the semantics of the diagram, preserving the geometric relationships dictated by the visual language. Our research has particular application to user interfaces for pen-based computers.