Response to Artistic Data Visualization: Beyond Visual Analytics by Fernanda B. Viégas and Martin Wattenberg

Fernanda Viegas and Martin Wattenberg explored the artistic practice of infovis, otherwise known as information/data visualization in their article entitled Artistic Data Visualization: Beyond Visual Analytics. After making clear distinctions between artistic infovis (representation of data with the intention of creating art) and scientific infovis (representation of data with as impartial a view as possible), these authors have then pondered the following question at the conclusion of their article:

“Should data visualization researchers investigate ways to support making a point, as well as disinterested analysis?” (Viegas and Wattenberg, 10)

At first the very idea of merging artistic and scientific approaches seemed paradoxical to me; the very purpose of science is to impart no emotional or otherwise conscious bias onto the represented data, whereas artistic infovis is its antithesis in the sense that art’s purpose is to represent data with an agenda, and an emotional attachment or motivation. How could making biased data sets be a good thing in terms of research, and why would one want to ‘cloud’ a data set with such bias?

But the more I thought about it, the more I realized that in order to transmit relevant data, you need a point, and without a point, the data looses its relevance. A point, or purpose, reveals patterns in data and reveals connections that were previously imperceptible. For example, Jason Salavon’s piece Homes for Sale is not an impartial data visualization piece; the images he used to generate ‘average’ representations of the middle-priced real estates found in certain geographical areas were highly modified before being fed to an algorithm for output. Presumably, he had to scale each image so that they were all the same pixel dimensions, and Viegas and Wattenberg asserted that he purposely overlapped his images so that the similar components of his images would line up.

Since he went through the trouble of ‘tampering’ (resizing and overlaying) with his ‘pure’ data ( the found images of real estate) the output images we get as a result are striking in the amount of data that they communicate. Viegas and Wattenberg describe the information displayed by Salavon’s series of aggregate images well, surmising that “...Miami boasts the bluest sky whereas Dallas has the greenest grass. Seattle, on the other hand, seems awash in an assortment of gloomy grays” (Viegas and Wattenberg, 4).

If Salavon had not done this, then the ‘average look’ of the places he was cataloguing would have been lost; ground, house, and sky would have overlapped at random, and the data would have been abstracted instead of made clear.

Is not the very purpose of data visualization (even in a scientific, research-based context) to represent data clearly and concisely? Then maybe using an artistic sensibility when creating information visualizations is not such a bad thing. Artistic intention seems to be the best way of ‘parsing’ the raw data that often finds its way into completed scientific data visualizations. Even Golan Levin’s piece The Secret Life of Numbers, as minimalistic and clean as it is, uses some sense of design and artistry to create a user interface that is informative and reveals patterns in the data that no scientific graph could.

The answer to Viegas and Wattenberg’s question, although perplexing at first, has become rather clear to me. Through the use of clearly defined purpose and intention, data sets may become more intrinsic to the very data that they represent. In this way, I feel that datavis researchers would in fact strongly benefit from investigating ways of ‘making a point.’

Response to interactive essay Exploring Emergence by Mitchel Resnick and Brian Silverman

I found the topic of emergence really interesting. The whole concept that complex objects, patterns, and behaviours can form from "... simple interactions in ways that are surprising and counter-intuitive" (Resnick and Silverman) is something that seems very analogous to currently accepted theories about the origins of organic life. New media artist Ken Rinaldo has even made it part of his artistic practice to explore this relationship, asserting that the "...integration of the organic and electro-mechanical elements asserts a confluence and co-evolution between living and evolving technological material" (Rinaldo, Artist Statement).

This parallelism between emergent and biological systems is not lost on either Resnick or Silverman, who make allusions to Darwin and his theory of evolution in regards to emergent digital systems. The theory of evolution is not the only theory they make allusions to; they aptly and perhaps unwittingly describe the basis of chaos theory (basically, the butterfly effect) when merely describing the behavior of one of their example programs:

"Small variations in the initial configuration of the squares can lead to large changes in the resulting patterns" (Resnick and Silverman).

But what does this mean, in terms of new media practice? Obviously it is a popular subject for many in the new media field; artists like Daniel Shiffman experiment with flocking behavior, Ken Rinaldo creates autonomous telematic spiders, and artists/programmers in many fields still build off of the Braitenberg Vehicles, a thought experiment originally conceived by cyberneticist Valentino Braitenberg some years ago. Will code progress until it has become analogous with DNA, effectively replicating the biological/organic behaviours that we as programmers have so often sought to emulate with code?

I think that this is a conceivable possibility, but it is one that will take many, many years to achieve. Science fiction has already made escapades into this hypothetical realm; in my science fiction liberal we have discussed in length the nature of life and what makes one truly alive and living. The film Ghost in the Shell (Mamoru Oshii, 1995), a futuristic sci-fi anime of a hypothetical world in which cyborgs and high-tech gadgets abound, featured a character who was a digital 'program' called the Puppet Master. The Puppet Master had gained sentience and argued that he was a person and was alive because he could recognize his own existence. The Puppet Master recounts how over time, his acquirement of data from various assigned tasks and analyses lead to the emergent formation of his own consciousness, which his own programmers identified as a mere bug in his programming. William Gibson's seminal cyber-punk novel Neuromancer also dealth with this theme; his two characters Wintermute and Neuromancer were AIs that had gained sentience through deliberate design.

We certainly are not yet at that advanced level of coding concerning AI and emergent intelligent/emotional behaviours, but contemporary coders and new media artists are working their way there. Through complex behaviours interacting together, emergent intelligence could hypothetically evolve. Daniel Shiffman's self-proclaimed 'organic' pathfinding algorithms display this kind of emergent and evolving behaviour that is made possible through code.

Emergence is a lucrative new ground for media artists to explore, one that will yield unexpected and fascinating results. It is my hope that new media artists continue to explore this field of coding, and I hope personally to start experimenting with emergent systems in the near future as my coding proficiency gets higher.