Another motivation for using dynamic figure captions is to tailor the content of these captions to users needs and preferences. Such an adaptation is useful because the overall amount of information mentioned in a figure caption can become quite large. Adaptation facilities should enable users to control the content selection. Several parameters are particularly important for adaptation:
Figure 3: Adaptation of figure captions. In the left part, the content
selection is customized. In the right part, the user
selects objects for monitoring.
Content selection. The caption generation process can be adapted to what kind of information is to be presented and to the level of detail. The information selection option (see the middle left part of Figure 3) allows a user to request a notification when a presentation variable has been altered, or when geometrical aspects of the image have been manipulated.
In addition, a user may control the amount of information by specifying the level of detail (see the information level option in the top left part of Figure 3). Selecting short figure captions means that only the most important changes in each selected category are presented whereas long figure captions inform the user about every modification to the selected property.
Object monitoring. Users should be able to express their interest in specific objects or regions (parts of the underlying model). Based on such a specification, the user will be informed about all changes that affect the objects or regions traced and of their current state (see the select tracing items option in the right part of Figure 3). For instance, if an object becomes hidden, the system comments on the visibility of such objects (e.g.\ ``the traced [object1] is currently hidden by [object2] and [object3]''). This is similar to debugging tools which allow the user to monitor certain variables. The selection can be performed either by picking the object in the model or by selecting an object via its name (see the list of objects provided in the right part of Figure 3).