Developing MORROIS (Mapping of Romance Realms and Other Imagined Spaces): Digitizing Geographic Data Drawn from Literary Sources

Place and space are critical elements of medieval popular romance, both in the journeys undertaken by the romance protagonist (and imaginatively recreated by the romance reader), and in the transmission of texts across space and time. These phenomena have driven critical interest in spatial markers in literary texts. The Morrois (Mapping of Romance Realms and Other Imagined Spaces) project, a digital geographic concordance of literary spaces, collects line-by-line instances of explicitly geographic place-name usage in Middle English manuscripts. The end goal of Morrois is to explore the research possibilities afforded through distant reading and various data visualizations (including GIS). Our poster will present data migration from Omeka Classic to RDF.
In its current form, Morrois uses the Omeka Classic content management system (encoded using Dublin Core metadata schema), and maps the places associated with the Alliterative Morte Arthure (found in the Lincoln Cathedral Library MS 91, “the Thornton MS”) through the Neatline framework. Since the beginning of this academic year until August 2020, we will be migrating this and other datasets to a new triple-store database platform with more nuanced and transferrable metadata schema. The rest of the texts from the Lincoln Thornton MS and Edinburgh, National Library of Scotland Advocates’ MS 19.2.1 (“the Auchinleck MS”) will also be added to the database, and we hope to offer a series of new visualizations via SPARQL Queries, as well as a more robust GIS that will allow for fuzzy spaces, a critical element of medieval geodata.
Our poster highlights our methodology, including the selection and customization of flexible and transferable ontologies and the benefits of RDF metadata modelling. We address some of the challenges inherent in data migration from a traditional relational database format to one geared for Linked Open Data. In the case of Morrois, Omeka Classic’s implementation of Dublin Core was effective for information about the manuscript texts themselves (the metatextual data), but unsuited for the line-by-line references found within the texts (the intratextual data). Migrating data from this “shoehorned” format to more accurate schema required considerable overhaul. Finally, we will address the current state of Middle English manuscript studies, wherein texts can be found in editions preserved in simple HTML format, printed and non-digitized critical and diplomatic editions, or in hard copy or digital manuscript facsimile. This situation means that data collection must follow different methodologies and different forms of data storage, including CSV or TEI.