I created a map displaying the locations of photos shot on rolls 1-5 of Charles Cushman’s film in the Charles W. Cushman Photograph Collection.
The map has a legend on the left which shows the colors that correspond to each roll of film that I mapped. For instance, all of the red dots indicate the location of photographs taken on the first roll of film in the Cushman Photograph Collection.
Clicking on a dot will give you some information about the photo: I didn’t include all of the information listed about each photo in the collection, but I chose the attributes which I though would be most relevant. First you can access a URL which shows you the photo. Then you can see the photo’s archive number in relation to the Cushman Photograph Collection, the roll the photo came from (this is also obvious in the color of the point), the photo’s frame number (how far into the roll of film was this photo taken?), the date it was taken on, Cushman’s description of the photo on both the film itself and in his notebook, and finally the address and lat/long of the photo’s location. It’s really important to note that some points (aka locations) include the details of several photos, since Cushman sometimes took multiple photos in one spot — if this is the case for any point, you are able to scroll through the point’s multiple entries using arrows at the top of the details box, although I wish the controls for this were a bit more prominent.
My source was the Charles W. Cushman Photograph Collection. The organization asked that I cite them in this way:
Metadata from the Charles W. Cushman Photograph Collection was created under the direction of the Indiana University Archives and published by the Indiana University Libraries.
I first spent some time exploring the Cushman dataset to see which attributes of the collection interested me. I became interested when I realized that the set had a column denoting specific rolls of film, because this attribute seemed to tie the data conceptually to the physical action and limitations of photographing the world. Once I’d identified the rolls of film as my main point of interest, I made five new spreadsheets and copied the collection’s data for each roll into a separate spreadsheet.
Then I had to decide which data about each roll I wanted to keep and what I wanted to delete. I ended up deleting a lot of the collection’s information for each piece because I thought it would be too overwhelming/irrelevant. I wanted people to see the photo, when it was taken, some information tying the photo to the collection in case they wanted to look for it in the collection’s own database, a brief description of the photo, and of course the address where it was taken. The rest of the information seemed superfluous for the purposes of this project, for example, I deleted columns with information about the photograph’s physical condition and the “genre” of the photograph.
Finally, I needed to prepare the spreadsheets to be geocoded into an ArcGIS map. The address for each photo took up about 20 columns in the spreadsheet because the Cushman collection had organized their addresses in this way — there was a column for street, street 2, street 3, country, county, city, etc. Many columns were blank. To prepare the spreadsheet, I used an amazing google sheets add-on called “Geocoding” by Awesome Table (thanks Austin for pointing me to this). This add-on allowed me to first concatenate the 20-ish address columns into one “full address” column. I did this for each spreadsheet, and then deleted the other address columns. Then I used the “geocode” feature of the add-on to add a latitude and longitude column to each location.
Finally, I exported each sheet as a CSV file. In ArcGIS, I imported each sheet as a layer, adjusting the colors to denote separate rolls of film. This aspect of the project was not very difficult, but I wanted to make sure I did everything cleanly, and created a simple-looking output so that people would be able to understand the map without too much explanation.
As I mentioned above, I spent a bit of time working on the presentation of the map, but I mostly just wanted to make sure that it looked simple. Then I created this WordPress subdomain of my main class blog site. Creating the subdomain was also simple, and I felt lucky that I happened across this really minimal, single-page WordPress theme. I wish I had found it when I was designing my class blog! I think it fits the presentation of this midterm well because it highlights the project and not the site. I changed the theme’s colors to look a bit like parchment or old paper, since this fits the aesthetic of the ArcGIS embed.
To embed the map, I used the same method of custom HTML as we used in class previously. I also made sure to include a legend in the embed of my map; this is key because it shows which colors correspond to which rolls of film.
I’m actually surprised about the significance of this project. I didn’t think it would be possible for me to create a very significant DH project for this midterm because of time constraints, option constraints, etc. And this isn’t a very significant DH project. But I think it’s a pretty significant DH project!
Because rolls of film have a temporality (all photos on one roll were taken one-after-the-other within the same time frame) and physicality (all photos on one roll were taken within a single journey from place to place) associated with them, visualizing the locations of photos on a roll of film helps you understand the journey that a photographer like Cushman was taking as he documented life in the United States. You can see some of the paths he took, and even the places he revisited. Something which interests me is how Cushman seemed to have been following the Wabash River while photographing on Roll 4 (blue). I think a next step to strengthen the project would be drawing lines between points based on photograph’s frame numbers for each roll. This could show you the exact succession of photos on a roll, and as a result, a more exact version of Cushman’s journey.