[openrisk]

> Building and querying the quadtree is intrinsically heirarchical

Glad to see I am not the only one having problems with hierarchy.

Interesting work at many levels (no, no pun): starting with the bluesky data availability, the processing and the visualization algorithms.

But its not quite clear where to place these visualizations in the data science spectrum. Conventional numerical graphics have (over time) developed a sophisticated grammar that allows fairly precise reasoning and inference. So they are heavily used in scientific publications, in the financial sector etc. for real information transmission (People might even reverse engineer a plot to recover data!).

With networks and graphs, besides a general feel for the topology / connectivity or clustering its kinda hard to pin down what is the transmitted information. Not clear if useful grammars covering such large graphs are yet to be invented or if this is the nature of the beast.