Hi, everyone! My name is Hillary Brady, I am a second year in the Public Humanities Master’s program! Through this program, I have become particularly interested in digital humanities work. While I have a background in writing for an online audience, I am excited to learn more about how to best represent stories, history, artifacts, etc. in a more visually engaging way!
As such, Lev Manovich’s “What is Visualization?” introductory article was helpful to me, in orienting myself in the basics of how to think about visual representation.
While Manovich’s article gives a wide overview of the distinctions between different forms of visual representation, the article’s primary issue is the concept of spatial relations being the biggest signifier to graphically represent change. Our most basic graphic visualizations (line graphs, bar graphs) rely on simple geometry—the use of “spatial variables,” primarily position, size and shape, to represent data. This is a way humans inherently view things, Manovich argues, which helps create a more concrete way to view discrete data. It creates a structure that its readers are comfortable relating to in otherwise large, unstructured data.
As such, secondary aesthetics, such as color or symbols, are used to represent less important or second tier information, while spatial relationships are used to represent the “most important dimensions.” (With a few exceptions that prove the rule: for example, a stop light uses color to convey the most important information, but only works because the lights are all the same size, throwing out spatial relationships as a signifier).
However, I wonder if this might be too simplistic. A good project to look at is The New York Times’ “Reshaping New York” (http://www.nytimes.com/newsgraphics/2013/08/18/reshaping-new-york/). It shows new buildings constructed under Mayor Bloomberg, indicated on a topographical map using color—however, it is clear that not all buildings are the same size. As such, it uses color as the primary signifier without discounting spatial relationships (on the map, spatial relationships are actually fairly important—it shows target neighborhoods for construction, for example).
Another New York Times project that might use color and space in a similar way is the “Inaugural Words” project (http://www.nytimes.com/interactive/2009/01/17/washington/20090117_ADDRESSES.html?_r=1&) that makes word clouds out of Presidential inaugural address transcripts. It might be interesting (at least for more contemporary addresses), to track party rhetoric in presidential speeches—what type of language does a Republic use, versus a Democrat? Since color is such a weighty and common signifier in modern-day American politics, combining color with a separate word cloud might bring new information out of the data set.