Liz's Project Details


What You Get Is What You See: Knowledge Production in Visualizations

By: Liz Shayne, Ludic Analytics Team

 

Goals:

As part of the broader work of the Ludic Analytics collective, this project aims to reexamine the roles that visualizations play in uncovering and creating knowledge around a text. The images that accompany digital textual analysis are rarely the subject of critical inquiry other than as adjuncts that support an article’s argument. This project proposes to focus on what the visualizations themselves are saying and, rather than accept their roles as subsidiary displays of knowledge produced in a computer and analyzed in an article, think about the images as ways of knowing in their own right. In order to move away from thinking of visualizations in a supporting role, this project will experiment with a multiplicity of visual techniques and allow the results of the visualizations to structure my inquiries. While “playing around” with multiple kinds of visualizations, this project will explore an epistemology of visualization, looking at the knowledges they instantiate as well as those which they occlude, and create a framework with which to think about the transformation that happens to knowledge as it becomes visual. As part of that process, I will engage with questions regarding what makes a visualization valuable both as an aesthetic image and as an interpretive tool.

 

Methods:

Though this project is an attempt to think broadly about visualizations, I required a single text through which to approach the multiple possible forms of visualization. I chose George Eliot’s Daniel Deronda (henceforth DD), a familiar text from prior research that is long enough to support many different kinds of analyses. Though most novel-length works are amenable to being visualized, Eliot’s novel is particularly suited for visualization work: the plot revolves around two protagonists, Gwendolen Harcourt and Daniel Deronda, whose lives intersect at various moments, but who still inhabit their own narratives. Depending on whether DD is treated as a single text or as a pair of joined narratives, different styles of visualizations become possible.

 

This project’s approach towards visualization is somewhat whimsical, relying at it does on semi-directed experimentation that first creates simple images and then uses those results to inspire more goal-oriented and complicated visualizations. Rather than create a set of rules to dictate my interactions with the text, I rely on the visualizations themselves to suggest the direction of future visualizations. This form of textual investigation is inherently playful and is a relative of the approach to reading that Stephen Ramsay calls “Screwing Around” in his talk about the hermeneutics thereof. The focus in this experiment is on the process rather than on the results. As Jentery Sayers observed about work in the Digital Humanities, “Everything is iterative. Our process is our product” (Cenki, #DHSI Tweet). In this spirit, we started the blog Ludic Analytics to document our respective progresses and processes. The material on the blog serves as a set of guidelines for how the group saw fit to encounter and engage with visualizations in a playful and evocative manner.

 

Results:

As noted earlier, in digital work, the process is synonymous with the results. This section will therefore be presented as a cross between a narrative and a curatorial exercise. It will consist of three visualizations—from the beginning, middle and end of the project—tied together by an explanation of how each one influenced this project’s conception of what visualizations do and what forms of knowledge come from each. Though more than a dozen images were produced over the course of this project, these three are representative of the three broader types of visualizations that I createdthose based on text analysis, artistic endeavor and network graphingand they are capable of speaking for the remainders. A full record of all the visualizations associated with this project can be found on the Ludic Analytics blog, along with a more detailed discussion of how they fit in to the investigatory trajectory of this project. Additionally, each visualization below will be followed by a link to the particular blog post in which it appears, to aid in contextualization.

 

Liz's Visualizations

 

 

Conclusion and Future Directions:

Though the final results of this project have answered the original question by elucidating how the process of creating visualizations is the process through which the visualizations instantiate knowledge, an obvious problem remains. As conceptualized here, visualizations are always infinitely more useful to their creators than to their audiences. And while the creator can certainly provide information surrounding a visualization that explains how she interprets the image, the knowledge will never be anywhere near as complete for the viewer as it was for the creator. As I conceive it, the next step in this project is to explore methods of conveying the knowledge that comes from visualization-making through the visualization itself. One possible way to do so would be to focus more on aesthetically pleasing visualizations that invite the viewer in rather than pushing him away. These visualizations were made with aesthetic principles in mind, but this project was not focused on how beautiful visualizations could be. More work in creating beautiful images that inspire interaction could be fruitful. Another direction would be to investigate dynamic visualizations that quite literally require the viewer’s participation. Dynamic visualizations could convey more information, such as the temporal nature of narrative that routinely disappears into the two-dimensional surface of the static image, and also could provide an interface for the viewer to produce and discover knowledge in a manner similar to the creator. This could be the future of the playful visualization, a visualization that actually invites one to play with it.

 

Works Cited and Resources:

 

Eliot, George. Daniel Deronda. Project Gutenberg.com. 2010. Originally published 1876. http://www.gutenberg.org/cache/epub/7469/pg7469.html. Web.

 

Hayles, N. Katherine. How We Think: Digital Media and Contemporary Technogenesis. Chicago: University of Chicago Press. 2012. Print.

 

Hoover, David. "Using the Craig Zeta Spreadsheet." David L. Hoover at NYU. 2012. https://files.nyu.edu/dh3/public/UsingtheCraigZetaSpreadsheet.html. Web.

 

Lima, Manuel. Visual Complexity: Mapping Patterns of Information. New York: Princeton Architectural Press. 2011. Print.

 

Ramsay, Stephen.  “The Hermeneutics of Screwing Around; or What You Do with a Million Books.” PastPlay Conference. Vancouver: University of British Columbia Press, 2012. http://www.playingwithhistory.com/www.playingwithhistory.com/wp-content/uploads/Hermeneutics.pdf. Web.

 

Pavel Cenki (@placesense). “everything is iterative...or...process has become our product @jenterysayers at #digped #dhsi2012.” 5 June 2012, 1:26 pm. Tweet.