Pasts in Space: Two workshops on historical gazetteers
Gabriel Bodard, Reader in Digital Classics, writes about two training workshops held on 21 and 22 October 2024
This month, we were visited by two eminent scholars who specialise in the digital treatment of classical and historical place data, for a total of five hours of tuition, discussion and exercises around historical gazetteers and research on ancient places. Dr. Valeria Vitale is a Lecturer in Digital Humanities at the University of Sheffield, and a contributor to the Machines Reading Maps project; she was previously based here at the ICS while working on the Pelagios Network, and her expertise also includes Linked Open Data and 3D Modelling. Dr. Tom Elliott is the Associate Director for Digital Programs and Senior Research Scholar at the Institute for the Study of the Ancient World, New York University, and director of the Pleiades Project; he also works in ancient geography, Linked Open Data and Latin epigraphy, and is a leading member of the EpiDoc community.
Pasts in Space was a two-day series of workshops for historians, archivists, and archaeologists. Through discussion and hands-on exploration, participants were invited to learn about resources and techniques for using, creating, and sharing information about past places and spaces for their research, teaching, and scholarly communication. Particular emphasis was placed on theoretical discussion: how to deal with incomplete, unlocatable, contingent, disputed and uncertain toponyms and locations over the course of a project or investigation.
The sessions, held in the Senate House MakerSpace, were fully booked and extremely well attended, with a mix of students and researchers at all levels of experience, from many disciplinary backgrounds, and the length of the UK, including library and museum archivists. Many had specific projects of their own that the workshops helped them in the planning or execution of, and all were very engaged and responsive, asking questions, sharing reactions and examples, and participating enthusiastically in exercises and discussion.
On Monday October 21 (event 1), we focussed on getting started with the Pleiades gazetteer of ancient places. The first part of the afternoon was spent talking about finding and using Pleiades data, outlining the nature, structure, and extent of data in the gazetteer, and showing in detail how to search, collect, and download place information. After a much needed break, we turned to practical and hands-on treatment of creating and changing Pleiades data, including editing guidelines and review and editorial workflow. Registered users on the Pleiades site (and anyone can request an account with some explanation of interest and credentials) are able to modify descriptive summaries, toponyms and connections between places; add more accurate or more complete spatial geometries; contribute notice of new bibliography; or craft new entries for places and spaces not yet represented in the gazetteer.
There was some live editing of Pleiades entries in the room, but most participants chose to contribute via discussion with the instructors, suggesting sites in need of improved documentation, geometry or relationships, and asking questions about the scope and purview of the gazetteer. Can we include fictional or mythological places, we asked? We explored what a Pleiades entry for Hades would look like, and one of us even created a new entry for Plato’s Atlantis (not yet approved, but when it is, this link will be where you can find it!)
On Tuesday October 22 (event 2), we worked on getting started creating and curating your own gazetteer. For the first couple hours, we discussed the concepts of usable geographical data, including spatial knowledge systems and models for understanding and structuring spatial data, from coordinates, names, descriptions to relationships. Places without coordinates (either because unknown, uncertain, unstable, or just not of interest) are perfectly feasible features of a gazetteer, unlike a GIS. Gazetteers may record different definitions of the same "place." We also discussed machine-actionable formats for past-oriented data, including the Linked Places Format, the Text Encoding Initiative, and comma-separated values (CSV or spreadsheets). After the mid-afternoon break, we moved to more collaborative and engaged discussion of creating and curating place data, including considerations of incremental data collection and editing, and the importance of tracing origins and controlling data versions over time.
Many of the participants in this session had their own projects, datasets, GIS or maps, which led to very different models of what data and metadata were essential, important or superfluous in a table of ancient places or placenames. Discussion was lively and fruitful, and sometimes surprising—cries of “But how would that work?” or “That’s not what a ‘place’ means in my project!” were occasionally heard. It was clear that ideas from this workshop were already being incorporated into existing and new projects, that thinking had been changed and enriched, and relationships made between researchers and datasets.
The instructors, and other colleagues here at the ICS and IHR, very much look forward to hearing about progress and future developments of many of the projects we heard about this week.