Events Graphic
MIT Media Lab about us . academics . sponsors . research . publications . events . people . contact us
 

Dissertation Defense

WHAT:
Barbara Barry (Warner Brothers Fellow):
"Mindful Documentary"

WHEN:
Monday, April 11, 2005, 10:00 AM EST

WHERE:
Bartos Theatre, MIT Media Lab (E15)

DISSERTATION COMMITTEE:
Glorianna Davenport
Principal Research Associate
Director, Media Fabrics Group
MIT Media Lab

Walter Bender
Executive Director, Media Laboratory
Senior Research Scientist
Director, Electronic Publishing Group

Erik T. Mueller
Research Staff Member
IBM Research

ABSTRACT:
In the practice of documentary creation, a videographer performs an elaborate balancing act between observing the world, deciding what to record, and understanding the implications of the recorded material, all with respect to their primary goal of story construction. This thesis presents mindful documentary, a model of a videographer's cyclical process of thinking and constructing during a documentary production. The purpose of this model is to better understand how to support documentary creation through systems that assist the documentary videographer in discovering new methods of observation, ways of thinking, and novel stories while recording the world.

Based on the mindful documentary model, a partnership between the videographer and a camera with commonsense-reasoning abilities is established during capture and organization of documentary video collections. Knowledge is solicited from the videographer at the point of capture; it is used to generate narrative or contextual shot suggestions, which provide alternative recording path ideas for the videographer. Thus the system encourages the videographer to reflect on the story possibilities of a documentary collection during real-time capture.

Qualitative results of studies with a group of videographers—including novices and experts—showed a willingness to take suggestions during documentary production and, in some cases, to alter the recording path after reflection on shot possibilities presented by the system. Moreover, suggestions often had increased influence on the recording path if they were not taken as directives but as catalysts, i.e., prompts to expand thinking about the documentary subject rather than explicit shot instructions.

Critical lessons were learned about methodology and system design for documentary production. As a documentary is built, evidence of what the videographer has learned is represented in the documentary. The model, methodology and system presented in this thesis provide a basis for understanding how videographers think during documentary construction and how machines with commonsense-reasoning resources can serve as creative storytelling partners.


MIT Media Laboratory Home Page | Events Main Index