project

This project description is adapted from the most recent Seeing Syllabi project narrative. Thanks to Pamella Lach, manager of the Digital Innovation Lab, for co-authoring; I’ve revised and made some additions for this blog. I am grateful to the Institute for the Arts and Humanities and the DIL for providing the freedom of a fellowship semester in which to pursue this project.

The idea for the Seeing Syllabus project grew out of brainstorming sessions with some colleagues (Steve Brauer, Christine Tulley, Zac Zimmer, and Michael Simeone, any of whom may rejoin our efforts at some point) at the Digital Humanities High Performance Computing Collaboratory, an NEH-funded summer workshop hosted between the University of Illinois at Urbana-Champaign and the University of South Carolina at Columbia. With an initial goal of extending the functionality of existing course and content management systems, I originally envisioned a suite of tools that would incorporate a searchable and analytics-ready corpus of syllabi into a robust and collaborative course creation system. Two linked features were included in the initial proposal: 1) a graphical, unit-based course creation and scheduling tool that enables student and instructor inputs and 2) a tool to produce interactive visualizations of data drawn from significant corpora of syllabi and other course documents.

With Pam’s help and the support of the Digital Innovation Lab and the Institute for the Arts and Humanities here at UNC-Chapel Hill, I’ve worked to scale the project to semester-long fellowship size–and in the process, I believe we’ve uncovered the truest intent and widest potential utility at the heart of the original concept. As a DIL/IAH Faculty Fellow, I will focus on developing a core and foundational feature of the original proposal: namely, visualizations of content (metadata) mined from a large collection of syllabi spanning many disciplines and fields of study.

The resulting prototype will comprise a web-based graphical interface to the syllabus metadata that allows users to query the database and produce dynamic, interactive visualizations of their search results. These visualizations will allow users to make and explore visual connections between syllabi across the disciplines. The Seeing Syllabi team and I believe that this project has generalizable implications, given its ability to interact with large sets of unstructured data (a goal of and significant challenge for many DH projects). Moreover, we anticipate that this prototype will be extensible, allowing us to integrate it into a more elaborated course creation toolkit as we scale and pursue additional phases of the original vision.

We will develop the prototype of this visualization tool in loose collaboration with the Open Syllabus Project (OSP) at Columbia University. The current size of their test set is nearly 7,000 syllabi, which is somewhat small for the purposes of machine learning processes but is sufficient for harvesting enough metadata to power some basic visualizations.

I believe that access to a searchable and analyzable corpus of syllabi will be useful to a wide audience: education researchers and historians, instructors seeking inspiration, students seeking courses and cross-disciplinary collaborators, the generally curious, and more. How would you use it? We’re open to input and always seeking collaborators. Get involved! E-mail me at tnicholas@cs.unc.edu.