Structures of Thought:  Common Readers and Data Modeling Annotations

1. Abstract

Common Readers is a digital initiative dedicated to gathering and analyzing annotations in early modern playbooks through a custom-designed relational database. The corpus of the pilot phase consists of drama in English, 1550-1660. The goal is to foster a community of researchers to contribute to a growing dataset of annotations that will, ultimately, be part of a public-facing site. To date, there have been two major trends in annotation scholarship: the study of remarkable men (e.g. John Dee) and the study of remarkable books (e.g. a copy of Shakespeare’s First Folio held at Meisei University). DH work on annotations has followed suit, striving to make annotations by particular readers in particular books easily searchable.
Common Readers stands in contrast to both these scholarly trends in its potential to illuminate the reading habits of the less extraordinary, anonymous men and women who read, marked, and commented upon texts. This talk will introduce the audience to Common Readers and solicit feedback on the existing annotations data module to improve its implementation, accessibility, and appeal to a variety of users. It considers the construction of the data schema through particular use cases as an opportunity to reflect on the process of structuring data, and in particular on the practical and scholarly implications of the choice between an hierarchical encoding structure like TEI and a relational database. (For an example of the former, see the Archaeology of Reading project; for an example of the latter see Derrida’s Margins.)
Moreover, this talk will consider the way in which annotated books--as multivalent, hybrid objects--challenge many traditional structuring principles in and outside of DH. As an inquiry into the best approach to annotation studies, Common Readers raises broader questions about the ontology of annotated documents that engage many disciplines and subfields, offering potential applications for researchers who are curating data about complex material objects, designing platforms for capturing annotations, or practicing scholarship related to object-oriented ontology.
In this emphasis on data structures, Common Readers seeks community guidance on its approach to making highly particular data not only publicly shareable but widely useful. Originally, the plan was to provide the (future) public site with an export function, allowing users to work with project data in whatever way best suited their needs, and to deposit versioned datasets for general use in a repository like Zenodo. Both practices are likely to benefit other early modernists, but probably not a general audience. While the appeal of Linked Open Data (LOD) is strong) the best way in which to structure such highly specific data (derived from a very particular use case) in a way that would comply with recognized standards and be widely machine-readable remains unclear. The project team welcomes advice in establishing these best practices, as well as for redesigning the annotations data module itself.
Addendum: The disruption caused by COVID-19 has delayed finalizing the annotations data module discussed above. If you would be interested in reviewing it in the future, please email Rebecca Munson (rmunson@princeton.edu).