Language of Technology

Expressive Computation Lab
Processing, Wikifier, Python


Ars Electronica is an annual exhibition over the last 30 plus years where "artists, scientists, developers, designers, entrepreneurs, and activists" submit work guided by the philosophy of "never asking what technologies can or will be able to do, but always what it should do for us." The Association of Computing Machinery, on the other hand, is the world's largest computing society and strives "to raise awareness of computing's important technical, educational, and social issues around the world." Visualizing how both of these communities explore and investigate similar technologies and ideas, this system strives to show how media artists and technical researchers have discussed topics over time.


Both these resources, AE and ACM, have digital libraries that contain information of submitted projects or papers over time. The AE archive includes artist descriptions for each awarded submission from 1987 to 2019. ACM's digital archive, which is much larger than AE, contains all the papers within the association. In order to extract this information from both archives, the websites were digitally scraped and converted into two different databases.


With two different sets of texts it was crucial to find a way to properly compare these works. UCSB's digital humanities department has been working on a tool called the Wikifier, which made this comparison possible. The Wikifier is a tool that reads in the given text and then is able to select certain keywords or phrases and match them to a corresponding Wikipedia title. The Wikipedia titles act as a "third party system," which is able to connect these two different texts to one source, Wikipedia.

Running the Wikifier analysis led to a list of topic words (Wikipedia titles) for each database. After removing the non-overlapping topic words, it was now possible to dig deeper and learn how each topic word was being discussed from each database and how it changed over time. Using a collocation method, one can understand how the most frequent words being used to discuss the topic and how each database was discussing the topics similarly or differently.


With this unique data, new visualizations are created where both the data and the visual are of equal importance. This structure allows users to understand the history of topics as well as further learn how each topic changed over time. As the user begins analyzing the visualization and understanding the connections to history it is making, then the user can open up particular ACM papers or AE descriptions and better understand how these texts arrived at their particular moment in history.

The previous details on the visualization are innovative in the fact, that no other visualizations are able to show these trends in history over time when looking at a text based system comparing two different datasets. This innovation leads the user to understanding topics over time and being able to use a tool where specifically new comers to Ars Electronica will not be able to use the current implemented archive, but can use a tool like this and initially start by looking at things they are interested and then link to the particular AE projects and gain a perspective of how AE has changed over the last 30 years in comparison to ACM.