Fisheye views show the point of interest in detail and the overview of the graph in the same window. This is done by distorting the picture. The picture is scaled nonuniformly. Objects far away from the focus point are shrunk while objects near the focus point are magnified. The degree of visual distortion depends on the distance from the focus point. The visual effect is very similar to the fisheye lenses in photography (Fig. 32, right).
Fisheye views were inspected by [Fu86, SaBr94, No93, KRB95, MiSu91, FoKe96, StMu96, CCFS96]. They can be divided into graphical fisheye views, where the distance from the focus point is a function of the coordinates (e.g. the Euclidean distance), and logical fisheye views, where the distance is any logical function wrt. the graph (e.g. the length of the shortest path between focus point and node). A fisheye view might be distorting, i.e. objects far away from the focus are shrunk, and filtering, i.e. unimportant objects far away from the focus point are hidden. Further, a fisheye view is layout independent [No93], if first the demagnification or filtering is calculated and then the layout is done. Otherwise, it is layout dependent. Layout independent fisheye views have the advantage that the layout can be calculated using the knowledge which nodes are shrunk or filtered. This resembles the folding mechanism in that it saves space in the layout. Graphical fisheye views must be layout dependent, because in order to calculate the distance by coordinates, the layout must be known.
