I am interested in what it means to 'stay faithful to a text,' to accurately portray a text's 'true intentions' in translations and later editions.  If several different versions of a text were published during the author's lifetime, is it really accurate to count only the final version as the 'real' one to be cited and included in anthologies, essentially ignoring the preceding existences/lives of the text? This led me to use History Flow Visualization as a tool for observing and studying the ways in which Edgar Allan Poe's "Bridal Ballad" evolved over time through its different versions and revisions.  I used the versions from January 1837, July 1841, March 1843, August 1845, 1845, and 1849 (which can be found here):

I like the idea of considering every version as equally significant, suggesting that the 'real' or 'most faithful' version must somehow involve all of the text's lives.  With this interactive visualization, the reader/user can move through the different existences of the text and see the different revisions simultaneously -- and therefore has the ability to interpret the text based on a much more complete understanding of it than a single textual representation could provide.

In my other experimentation with visualization tools, I attempted to discover something new about either fairy tale rewritings or similar tales across cultures.  My most kind-of-successful experiment was using Voyant Tools to visualize the language of three versions (English translations) of "The Girl Without Hands" (German, Swahili, Italian):

I was most surprised by the prevalence of "king" in these texts, as well as by the amount of speech ('said', 'answered', asked', etc...) compared to some other fairy tales that revolve more around actions than dialogue between characters.