Clifford Beshers and Steven Feiner

Auto Visual

This text is partially taken from [BES93]

Beshers' and Feiner's AutoVisual designs interactive virtual worlds for visualizing and exploring multivariate relations.
The system designs n-Vision virtual worlds, which provides users with a 3D virtual world within which they can visualize and manipulate relations. Therefore AutoVisual uses worlds within worlds, an interactive method of representing multivariate relations with a hierarchy of nested heterogeneous coordinate systems (even the worlds). Each world may contain a graph encoding a subset of the relation encoded by its parent world. The subset is determined by the position of the world's origin relative to its parent. The user can grab each world using a Data Glove and move it throughout the space defined by its parent, thereby interactively exploring the data space.
An n-Vision virtual world is a hierarchy of interactors. Each interactor consists of four basic components: a set of encoding space, a set of encoding objects, a set of selections and a user interface. Each interactor has bindings for creating, copying and deleting child interactors.

• encoding space: this is the region where the graph is rendered. In general an encoding space may be of any dimension(up to three), arbitrarily positioned and oriented in the 3D world.
•  
• encoding objects: makes up the body of the graph
•  
• selection: determines a subset of the relation encoded by the parent.It consists of a geometric subset of the parent interactor's encoding space.
•  
• user interface: this is the user-interface component and it consists of bindings between the user's actions and the interactor's properties.

AutoVisual is a rule-based system, in which the user specifies the visualization task, rather than a particular visualization. The specification for one task will be performed on a set of relations. The whole task is the fusion of all these tasks, and AutoVisual must then create a visualization which will satisfy the overall task specification. The system makes a distinction between task operators and task selections. Task operators can be exploration, directed search and comparison, and they represent the user's fundametal cognitive operations when viewing and analysing a visualization. Task selections are subsets of interest for a particular task through simple constraints on the variables of a relation.
In AutoVisual's search strategy, each state represents a possible interactive visualization world and is a directed acyclic graph of interactors. The state-change operators simply add interactors or modify those already in the graph, producing the final visualization by incremental refinement. When it is necessary the systems backtracks to undo a design path that proves unsuccessful. Termination conditions are necessary. To provide termination conditions, AutoVisual uses two criteria to establish that it has found a satisfactory visualization: potential expressiveness and potential effectiveness. (These are Mackinlay's expressiveness and effectiveness criteria, modified for interactive visualization). A visualization is potentially expressive if it has the potential, under user control, to display all its assigned information over time. A visualization is potentially effective if over time it can present the information sufficiently clearly.

Visualization Concepts

Last modified on March 29, 1999, G. Scott Owen, owen@siggraph.org