Review of Practice

At present, Terror and Aesthetics renders 3D spheres in clusters that are staggered to causalities (in the terrorism dataset) and event intensity (in the state war dataset). They present at randomized locations on the screen and then proceed to slowly shift position while the lighting fades in and out. The intent here was to model an amorphous and shifting terrain within a menacing atmosphere that has no true temporal beginning or end. To that extent, the visualizations are successful. Building on this, future iterations will replace those spheres with near transparent renderings of human beings, so as to increase their emotional weight (a la the effectiveness of imagery from The Fallen of WWII, Poppy Fields, and Out of Sight, Out of Mind).

That the starting locations of these clusters changes with every reload, is meaningful and desirable insofar as it explicitly performs the non-self-identical nature of the text. However, the present extent to which it changes is too varied and sometimes results in the visualizations beginning with close to no objects on the screen. Future iterations will have to control for this while still maintaining this overall randomness.

With respect to the atmosphere, while the completely black surrounding environment contributes to the sense that there is no center, it may be possible and desirable to overlay semi-transparencies of relevant iconography (e.g. the symbols of terrorist groups, the flags of nation states, or visual synecdoches for the places affected (such as books to indicate schools, or the gender symbol for women to indicate abortion clinics)), while still maintaining this sense of spatial disorientation.

Finally, a deeply problematic part of Terror and Aesthetics is the inability to read incident information alongside the display. This absence decontextualizes what is happening on the screen, and cripples its ability to inform and thereby fully provoke. While I was able to isolate why this was happening – the text() function appears to be incompatible with the WEBGL rendering mode necessary for 3D graphics – I wasn’t able to figure out how to fix this problem. Future iterations will have to correct this, especially since this projects seeks to visualize readable comparison as it simultaneously provokes an emotional response.