This post describes my capstone project for the NEH-sponsored Institute on Digital Archaeology Method and Practice (#MSUdai) that I attended at Michigan State University in August. I’m cross-posting here from the institute’s website. You can read my application to the institute in a previous post on Documenting Cappadocia.

At its heart, the Digital Catalog of Cappadocian Ceiling Crosses will be a database of my dissertation data, deposited in open repositories and visualized for public dissemination as an interactive catalog of all known monumental ceiling crosses in Byzantine Cappadocia, a region that is now central Turkey. Cappadocian architecture offers one of the largest groups of late antique and medieval ceilings available for study because rock-hewn architecture is fairly durable, and dozens of monuments survive in the region. My dissertation’s underlying argument encourages readers to rethink the use of overhead space and monumental imagery: giant cross imagery placed overhead was designed to manipulate and inspire medieval Christian viewers within the spaces beneath.

The first deliverable from this MSUdai project will be an open data deposit into two repositories called Open Context and KORA. (A secondary contribution will be made to Pleiades, an ancient world gazetteer of places). The data will then receive public-facing visualizations based on the repositories’ built-in features. For instance, Open Context has a graph option, and KORA data can be visualized in a WordPress site using the MATRIX Connect plugin. (Theoretically, the data could be endlessly remixed or visualized in alternate formats such as Storymaps, Scalar, or mbira, but I’m taking Ethan’s advice to keep the scope creep at bay!)

Variations of this project have been in the ether for a while as I’ve learned how to think of art history research as reusable data. At LAWDI (a previous NEH-sponsored event called the Linked Ancient World Data Institute), I was able to recognize how this kind of project could take shape in the larger sphere of ancient world Linked Data, with a goal of a Pleiades contribution. At my home institution, several Graduate Center people and groups—the New Media Lab, Digital Initiatives programs (including the Digital Fellows training funds that helped me get to the institute), Mina Rees Librarians, classmates, and professors—have been incredibly generous with time, advice, and resources toward the digital project taking material shape. With their help, I’ve made some progress, but several areas of decision-making have been stalled.

Why, you might ask, would I need an archaeology institute to weigh in after all that good advice? The short answer is, it takes a village to link some data. A more nuanced answer would include the benefits of working with institute faculty members who not only know data, but who have collected and cleaned and archived archaeological data themselves, and who regularly think about the various levels of specificity (or lack thereof) in field recording. Another direct effect of MSUdai on this project the faculty’s knowledge of technologies, Content Management Systems, and platforms that are designed or have long been utilized for archaeological data.

The aha! moment for me in my new workflow came with the the solution of the long-term archiving of data with a (somewhat) automated workflow into its public-facing visualization, thanks to Eric Kansa’s suggestion. Before the institute, I kept trying to hack the WordPress content management system to use as a database, or to learn Drupal (for its more sophisticated construction data and metadata from the back-end) and then deposit the data somewhere for archiving. My new MSUdai approach is just the opposite: deposit the data in stable repositories, and then use the repositories’ built-in tools to visualize it into WordPress (using the MATRIX Connect plugin) or Open Context’s graphs feature that I can embed into a blog post. With this workflow, the data itself won’t be succeptible to updates (of browsers, platforms, software, etc.) from my end (as WordPress and Omeka are). The work of visualizating it will be simplified and enhanced by using tools that are specifically designed to draw from my chosen repositories.

In a nutshell, my first lesson learned from the capstone project is to start with a stable data deposit and then do all the “fancy” things with visualization. In business or grant-reporting, we’d call the data contribution my minimum viable product. My goal is to document and share my decisions and workflow so that this project can serve as a template for other researchers who would like to contribute small data sets to the wider open data ecosystem.

A.L. McMichael, @ByzCapp
PhD candidate in Art History
The Graduate Center of the City University of New York (CUNY)