I spent this past academic year working as a web archiving fellow with the New York Art Resources Consortium (NYARC). As part of that work, I was responsible for performing quality assurance on the archives of the Brooklyn Museum's website, so I became intimately familiar with the structure of the site and the metadata of its objects. Of course, I also just enjoyed looking at the art. So I was interested in finding other ways I could work with the collection this semester.
I worked with data from the museum's web archives for the practicum this semester, creating something of a data physicalization of a hyperlink network diagram. What that translates to is: With a pandemic underway, I was looking for a distraction, so I created a cross-stitch project that sort of resembles a digital visualization created in Gephi.
For my PFHC project, I hoped to share that kind of crafty distraction with others by creating a Twitter bot that would post cross-stitch patterns of objects in the Brooklyn Museum collection.
Working with the Brooklyn Museum API, the plan was to collect the metadata of objects that had images and were presumably not restricted by copyright (even though this project would be transformative, and therefore fair use, that helped limit the total number of objects I'd be working with—starting with nearly 93,000 objects, these limits got it down to about 45,000). I would download the images, transform them into cross-stitch patterns, and have the bot post images of the pattern, the color key, and some metadata about the object. I made some progress, but encountered a number of challenges, including formatting issues in the json file I generated with the API, getting stuck on multiple for loops, and struggling to push multiple images to Twitter using Tweepy.
But I'll continue to troubleshoot, and I hope to have the bot up and running soon for all those crafters out there in need of a little art-filled distraction.