Visual focusing and transition techniques in a treeviewer for Web information access

We discuss the visual focusing and transition techniques embodied in a Bellcore Java applet called AMIT (Animated Multiscale Interactive Treeviewer), whose purpose is to improve Web information access. AMIT integrates fisheye tree browsing with search and filtering techniques. Multiple foci are selected by direct manipulation, content-based search, variable scoring functions, and what we call node sharing. Focus is realized with two methods of user- or system-driven tree reduction and visual emphasis is accomplished with variable font scaling and color. All transitions are animated and deal with the problems of adding and deleting tree elements, changing font sizes, and shifts in scrolling position. We relate our work to a notion of cohesion in visual discourse.