Computational Support for Compelling Story Telling Sara H. Owsley, Kristian J. Hammond, David A. Shamma Intelligent Information Laboratory Northwestern University 2133 Sheridan Road, Room 3-320 Evanston, Illinois 60208 +1 (847) 467-6924 {sowsley, hammond, ayman}@cs.northwestern.edu Categories and Subject Descriptors J.5 [Arts and Humanities]: Arts, fine and performing; H.3.3 [In- formation Search and Retrieval]: Information filtering General Terms Human Factors 1. INTRODUCTION The Internet is a living, breathing reflection of who we are, what we think, and how we feel. The pages that make up the Web form the book of our contemporary life and culture. They are the ongoing and changing buzz of our world. The latest embodiment of this cultural reflection is found in blogs. Not only are blogs widespread, but they are incredibly dynamic, with hundreds updated per minute. The existence of millions of blogs on the web has resulted in more than the mere presence of millions of online journals, but rather, they generate a collective buzz around the events in the world. 2. BUZZ Buzz (shown in Figure 1) is a multimedia installation that exposes and explores the collective voice generated by blogs. Using the most popular searches of the moment, it finds the blogs that are re- flections of the questions of our time. The most popular searches are displayed on a central screen. The blogger is embodied by virtual actors, displayed on surrounding displays, who external- ize these monologues by reading them aloud. It exposes the opin- ions and experiences of people in the world relating to what we are searching for. As an example of a Buzz performance, Table 1 shows the buzz on the topic of the popular reality television show “Project Runway.” In the performance, the words “Project Runway” appear on the cen- tral screen, while the four actors contribute to the performance by reading these discovered stories (found in blogs) aloud, in turn [3]. The actors are attentive to each other by turning to face the actor currently speaking. Figure 1: An installation of Buzz in the Ford Engineering and Design Center at Northwestern University. The virtual actors are telling stories on the topic of Janis Joplin. Story telling and online communication have been externalized in several related installations. Of the more well-known, Listening Post [2] exposes content from thousands of chat rooms through an audio and visual display. Similar to Listening Post, Buzz ex- ternalizes online communication, though in a more semantically grounded manner. Like Buzz, Mateas’s Terminal Time [4] sought to tell stories, though its stories were grounded in a common sense knowledge base as opposed to live real world information. 3. TELLING COMPELLING STORIES Starting with the popular web searches of the day as a context, we enabled Buzz to discover stories by searching for blogs that re- lated to these topics. A first pass at building Buzz revealed that the content of blogs is incredibly wide ranging, but unfortunately often very dull, including entries about what the author is eating for lunch, or how to install a wireless router, or a list of the author’s 45 favorite ice cream flavors. We needed to give the system strategies for finding stories that were compelling and engaging to an audience. To do so, we define a sim- ple model for the aesthetic qualities of a compelling story. These qualities include but are not limited to: an interesting topic, emo- tionally charged, complete and of a length that holds the audience’s attention, at the right level of familiarity, and involving dramatic situations. We designed Buzz to find stories with these qualities.