OK, to start let me be clear about something: What you are seeing here isn’t the result of my current research. I didn’t spend months laboring over the collection of data rendered in these images, or days over its analysis. In contrast to the primary and secondary data that provides the foundation of my own research, the numbers behind these images are almost entirely computer-generated.
Apart from this, what do all of these images have in common?
A few things that might be surprising.
First, they were all drawn in … Microsoft Excel. An odd choice, one selected to prove a point—that we have an uncanny tendency to typecast tools and disregard their other capabilities. Ubiquitous desktop applications are a great example. It’s useful to demystify visualization tasks. Even complex ones are much more within reach for many (not all) applications than is often imagined.
Second, each of these images was constructed using simple trigonometric equations, with only slight modifications in each instance. Coupled with a little repeated computation (available in this case through loop structures in VBA), these extremely basic equations can go an impressive way. In this case, I’m tweaking only four inputs (a,b,c,d) and letting the computer run with it. The majority of the code used is provided for those interested.
Of course, there’s no reason to limit ourselves to two dimensions when exploring where basic mathematical building blocks, a little computer code, and available visual renderings can take us. Leveling up a bit, only a slight increase in code and applied trigonometry gets us into a much more interesting space. Here again I’m restricting myself to rendering the resulting visuals to a relatively limited environment, essentially using cells in a spreadsheet as I might pixels in an image. (Incidentally, to see this kind of thing taken to an extreme, see the work of Tatsuo Horiuchi: http://www.spoon-tamago.com/2013/05/28/tatsuo-horiuchi-excel-spreadsheet-artist/.) In this case, simple conditional formatting is used to apply colors to cells based on content, and a first step in generating a higher dimensional rendering is achieved, just barely, through the 3D bar chart option in Excel (not one I typically use, but one that provides an interesting perspective here).
But again, all of this is just a teaser. There’s a broader reason why I wanted to use a pseudo-pixilation approach to begin with. It’s a convenient means to an end, for test-casing graphical “formulae” that might be applied in a more compelling context.
And thanks to the gaming community (of which I am only really cognizant because of my son), we have many. As a result, these mathematical visual renderings need not reside in isolation. They can become the foundation of enhancements to entire virtual worlds (as presented below, where I’ve taken the math and programming into the world of Minecraft via Python). It’s not the Hagia Sophia, but, hey, it’s only a few lines of code. If a few lines of code and a little math can get you this far, imagine where some serious study in applied mathematics, computer science, and context-specific visualization can take you.