Triggered a great lost memory: When I was on a dev team testing the graphing systems for a scientific data presentation software back in the early 90s I would try to come up with "fictional data sets" that would make interesting graphs. I tried to make them pretty, or weird, or just break the systems. Occasionally, it was so interesting I'd end up looking at the data / formulae that lead to the graph / design. Great creative space to be in for jumping back and forth between math and art.