Photogrammar is a web-based data visualization of 170,000 photographs from 1935 to 1945 found at http://photogrammar.yale.edu/. It allows not only for visualizing, but also organizing and searching these photographs. This website was created by the United States Farm Security Administration and Office of War Information (FSA-OWI).
The sources for this project come from the FSA-OWI File collection of images. Sponsored by the federal government, the Farm Security Administration — Office of War Information was given the task of documenting America, which produced a plethora of photographs from many well known photographers. The negatives were sent to Washington, DC where they were stored in a collection that eventually came to be known as the FSA-OWI File.
The processing for this project began with the cataloguing of the collection. Paul Vanderbilt joined the fSA-OWI in 1942 and created an organizing system for the collection that included the Lot Number system and Classification Tags system for the photographs. The lot numbers signify a set of photographs organized around a shooting assignment, so they usually feature one photographs photos in a single place, and the classification tags have headings and subheadings that describe the subject matter of the photograph. This metadata allows users in Photogrammar to search through the collection. Apart from this, the Photogrammar team had to quantify the number of photos taken in each location and group them together by photographer, location, date, and classification in order to visualize it on their maps.
The presentation of this information is mainly done through maps. One of these maps shows the photographs by county, where the shade of color of the county represents the amount of photographs taken there. There is also a dot map that has different color dots for every different photographer. On these maps, you can zoom in and out, filter for a specific photographer, or adjust the time period that is shown. When you click on a county, it tells you the location and links to all the pictures taken in that county. It is important to note that the map only plots the approximately 90,000 photographs that have geographical information, whereas there is about 170,000 photographs total on record. These maps were created using leaflet, an open-source JavaScript library for mobile-friendly interactive maps, and CARTO, an open platform for location data.
There is also a section of the project called Photogrammar Labs where experimental visualization techniques are being used to represent these photographs, such as a treemap that traverses through the headings and sub-headings of the images, a metadata dashboard that represents the photographs in California with different charts and visualizations, and colorspace that allows the user to explore the photographs based on hue, saturation, and lightness.
Written by Risha Sanikommu
I also chose to analyze Photogrammar. This was an extremely unique and interesting website, and I have never seen anything like it. I love your blog- there were some details you included that I didn’t even realize. For example, you mentioned that the color of the counties on the map represented the amount of photographs from there. I did not realize this and remember wondering why each county was pictured in the shade of green that it was. Excellent observation. I also love your detail about the map only showing 90,000 of the 170,000 photographs on file. I wonder why all the photographs aren’t on the website. Perhaps some were lost or deemed inappropriate. These photographs may have valuable information that is lost in Photogrammar.
I didn’t immediately realize the information represented by color and the relationship to the amount of photographs, either. I think the aesthetic choice of presentation and color is interesting in the context of the subject matter being represented.