Add this event to your calendar
“The whole endeavor of collecting images, categorizing them, and labeling them is itself a form of politics, filled with questions about who gets to decide what images mean and what kinds of social and political work those representations perform” - Kate Crawford and Trevor Paglen, Excavating AI. The Politics of Images in Machine Learning Training Sets
Session facilitators: Mariana Fossatti and Sunshine Fionah Komusana, Whose Knowledge
For the last 7 years through the #VisibleWikiWomen campaign, we have brought 27,000+ images of black, brown, trans, queer, Indigenous, and non-binary folks from the global majority to Wikimedia Commons, as part of our work to decolonize Wikipedia and the broader internet.
The thousands of images brought to Commons through the #VisibleWikiWomen campaign annually under a different theme allow us to experiment and better understand issues around the visual and data gender gap, accessibility and multilingual, automated tags, and biases in structured data. The structured data, however, does not tell the stories behind the images.
This session will draw on some of the images from the campaign to better understand what structured data is; how absent or skewed structured data is regarding images of women, and dedicate time to framing and adding decolonial structured narratives that tell stories in the plurality of our existence.
This is an online, introductory session open to Sussex humanities and social science researchers at all levels (please register using your @sussex email address). The session will be hosted on Zoom and the link will be shared after registration.
This workshop is part of the Digital Methods Accelerator, a series of open sessions hosted by the Sussex Digital Humanities Lab providing Humanities and Social Science researchers with critical digital skills.