Dec. 6 — Imagine a time when information was far less accessible than it is today. Without the power to pull out your smartphone and perform a fact check in a matter of seconds, you had to trust that the information presented to you was the accurate truth.
Throughout the 1930s and 1940s, Roy Stryker served as the head of the Information Division for the Farm Security Administration. He sent out photographers to capture images depicting rural life for a photo-documentary project. The total corpus of images consisted of over 250,000 photos, although only 175,000 of these were preserved and now reside in the Library of Congress. Yet many of these images never reached the public eye. Stryker would punch holes in the negatives of images he didn’t want the public to see, which is also known as “killing” an image.
Stryker used photography to change public perception and even Congressional perception of what was needed to alleviate the dire situation of farmers post-Dust Bowl and post-Great Depression. Whether or not you would consider this to be propaganda, it was certainly a media strategy which had significant impacts on our history. Now, with the help of XSEDE, the full stories can be told.
Co-PIs Elizabeth Wuerffel and Jeffrey Will of Valparaiso University, along with the rest of their team (Allen Craig of XSEDE, Marcus Slavenas of the National Center for Supercomputing Applications, and Paul Rodriguez of the San Diego Supercomputer Center [SDSC]) have spent the last year analyzing the images from Stryker’s photo-documentary. This research, funded solely by XSEDE and conducted on both SDSC’s Comet and Oasis, entailed using pre-existing algorithms, as well as creating new algorithms, to study the large corpus of images belonging to the Farm Security Administration – Office of War Information from 1935-1944, held by the Library of Congress. These images provide a glimpse into rural life and small-town America during this decade and throughout important events in America’s history such as the Dust Bowl and the Great Depression.
Their work serves many disciplines—the arts, humanities, American history, science, etc. By creating and expanding upon different algorithms, they’re empowering researchers to ask big questions, and relying on the power of supercomputer to help them answer these questions. By using a supercomputer to analyze this massive amount of images, it allows researchers to ask questions like “How did the images in this database depict rural life? What visual narratives appear across the corpus? Are there significant differences in approach from photographer to photographer? Which images did Stryker kill and what can we learn from those images?”
The entire article can be found here.
Source: Hannah Remmert, NCSA