Actually, this is the data.

Yesterday, I had the pleasure of meeting with Alison Langmead, the professor of Digital Studies and Methods 3000 (DSAM 3000). That is the doctoral seminar of which I am a part, and on account of which I have undertaken this project. It was an extremely helpful meeting, which of course is to say that it sent me back to the beginning of my work. I learned I had done it all wrong.

A pleasure of Dr. Langmead's classes is often that "wrong" is not really a thing, however. To the contrary, in her courses students are advised to be mindful of the decisions that they are making, of what comes from them and of what is learned in the process. In talking with her, I realized that I had leapt to the assumption that I would need to convert my PDF files from that format to XML, because I knew I would be doing "text mining" of some sort and knew that I couldn't perform such a fancy function over an image. Having done some very crude text analysis in Python and having worked from XML in doing so, I made a similar assumption that XML was the format I would need for this project.

As it turns out, Python might not be the best tool for my purposes, as to draw inferences from resources written in human-structured language it might be more beneficial to utilize topic modeling, as is supported by tools like MALLET. In that case, text files are best.

So, I have to go back and reconvert all of my files, this time from PDF to .txt. Stay tuned. I will repost here the original PDFs, as well as the .txt files as I have them ready. Reconverting is taking some time, so until this text is changed the collection of .txt files is representative of those that I have converted to-date.

Rejects in .pdf

Download PDF • 36.01MB

Rejects in .txt

Download TXT • 122KB

Colosseum flatware artwork, submitted for copyright by Oneida, Ltd. in March 2018.

Rejected for the second and final time in December 2019.