About hyperbolic tree
In a previous post I explained how the Bullet Graph and Sparkline widgets can help dashboard designers use valuable screen real-estate more economically. These charts are ideal for displaying single values and simple sequences of values. But a different set of problems arises when hierarchical data needs to be displayed, as users need to visualize not just the information relating to each individual node, but also the existence (and possibly the nature) of the relationships between nodes.
NetNow provides a wide range of widgets for displaying hierarchical data including trees, tree grids (such as the service summary) and multi-level pie charts. All of these are useful visualizations, but can become cluttered with very large trees.
A Hyperbolic Tree (or “Hypertree”) overcomes this limitation by reducing the screen space used to display nodes exponentially as the distance from the centre of the chart increases. This means that a Hypertree can pack much more information into the same amount of space than a standard tree graph:
The NetNow Hypertree widget enables easy navigation of very large hierarchies of nodes. Clicking on any node brings it immediately to the centre of the chart, automatically reducing the scale of surrounding nodes that are not the user’s current focus of attention.
The use of hyperbolic trees as a browsing mechanism for large hierarchies of data was pioneered by John Lamping, Ramana Rao and Peter Pirolli and is described in more detail in their original paper.