I was really inspired by MIT Media Lab Identity few years back. I liked the idea. Few days ago this Media Lab Identity thing sprung up on me, and I thought of a visual presentation scheme which would be aesthetically pleasing and also will be able to deliver a neat visual representation of each node in a tree like data set.
After thinking over a sleepless night, I sought this scheme out.
Let me explain.
In a tree-like data structure each node has a set of ancestors, going up to the top level (except for the top level nodes). In this version of the representation scheme, I scattered each node randomly over a grid and decided to render a single unique image for a single node, while each image containing not only the point of the node itself, but of it’s ancestors as well, straight up to the top level (except, again the top level nodes).
A dull image of a top level node,
While the test data set contained a 5 level tree, a third level node looks like this, depicting not only the third level (green) node, but also it’s ancestors (yellow and red).
You can also visualize relationships between nodes, such as,
In these two nodal images, you can see both these nodes belong to the same parent. (red, yellow, and green square is in the same positions while the blue changed). On the other hand,
These two have a great-grandfather type relationship.
The current version of the generator accepts the whole data-set at once in a tabular form and churns out the images. Written in Python & Processing.
Feel free to hack into it at https://github.com/rivalslayer/NodalIndentity. Also contains test data generator.