Selecting operations in an object-oriented user interface

A user can select actions in a user interface. An action can be selected by one or more "views" - for example, a "Print" action may be selected by selecting Print on a File menu, clicking on an icon of a printer, or pressing control-P. An action object defines, for each possible view representing the action, the attributes required to provide that view, and also an identifier to identify the operation to be invoked when the action is selected. The system identifies when a user selects a view and informs the action object of this. The action object then generates a command to invoke the desired operation. This mechanism achieves single event generation from multiple views, and therefore the application does not need to concern itself with how the user generated the call. The action views query the attributes of the action object directly, all the necessary data is held in one place, and with automatic view concurrency, the application developer no longer has to worry about keeping action views synchronised with each other.