Advancing digital literacy in an era of big data

    Text analysis tool developed by Geoffrey Rockwell and his team garners national award

    By Donna McKinnon on July 10, 2017

    What’s a philosopher doing dabbling in the digital arts? If you’re Geoffrey Rockwell, you’re developing cutting-edge computer technology that enables researchers to extract and interpret data from complex texts.

    The web-based text analysis tool, called Voyant Tools 2.0 and its companion volume, Hermeneutica: Computer-Assisted Interpretation in the Humanities, recently won the Canadian Society for Digital Humanities (CSDH) Outstanding Contribution Award, which is bestowed upon an outstanding project or publication by a Canadian researcher or a team based at a Canadian institution.

    Developed by Rockwell, Stéfan Sinclair (McGill) and their team, Voyant is simple to use and, according to Rockwell, has multiple functionalities. Once text is uploaded or pasted into a field, it can be analyzed in a number of ways, revealing features and relationships within the text or between texts. An example cited in a recent Compute Canada article involved a researcher using Voyant to identify references to “queenhood” in Shakespeare’s Henry VIII. The results revealed nuances of character identity previously missed, and though not an end in itself, advanced the research and posited new questions.

    The companion book, Hermeneutica: Computer-Assisted Interpretation in the Humanities, reflects – philosophically – on the Voyant Tools, encompassing the history of text analysis and the new research questions that arise out of this type of analysis. “Hermeneutics is an area of philosophy that crosses between philosophy and literature,” says Rockwell, adding that he and Sinclair share an interest in the philosophy of technology and how computers are changing interpretation.   

    “One of the things that stands out for me in this project is the collaboration,” says Rockwell. “We’ve been collaborating for over 10 years. And not just collaboration between two people, but trying to build those hybrid products, build the tools, see how it’s used by others, try it ourselves, write about it, present it. That sort of iteration is very different from what I was taught in graduate school. What I was taught to do was very solitary, meditative. Writing articles and books was the way that you shared it, not writing tools.” 

    Voyant 2.0 sees up to a million hits per month (with 200,000 unique visitors annually), and, thanks to the interactivity of the platform, is available in nine languages. Frequently used as a teaching tool in classrooms, its diverse range of research applications is also being used outside of the university setting in law offices, newsrooms and other organizations that deal with large amounts of data.      

    Rockwell says his interest in computer technology began as an undergrad, but it wasn’t until he became a teacher prior to graduate school that he discovered programming.

    “I was building little toys for students and trying to use the software that was out there, which at that time was on the Apple II,” he laughs. “In graduate school, I got an Apple Researcher internship. They were hiring graduate students as research assistants and I was paid to teach people how to use the Macintosh right at the moment when the web took off. I ended up getting a full-time job at the end of my graduate program working in instructional technology and visualization.”

    In 2008, the opportunity to teach at the graduate level and the vitality of the research climate at the University of Alberta brought Rockwell to the city [from McMaster]. Now a professor of philosophy and humanities computing, as well as director of the Kule Institute for Advanced Study (KIAS), Rockwell sees digital literacy as an essential tool in the practice of “democratic citizenship,” particularly in an era of big data and surveillance. 

    “More and more we’re seeing big data in various forms being used to manage us; essentially, replacing humans in the decision-making processes,” he says. “We all have to understand how we can be watched, and one of the things that people learn when they play with Voyant is the opportunities and limitations of big data. So just for democratic participation, as algorithms begin to replace people making decisions, we need to understand how those algorithms work. Playing with tools like Voyant is one of the ways people can understand, as opposed to just being told this is good or this is bad. You learn about cars by driving them.”