You are here

Embedded metadata and image retrieval: a critical case study

Vetusta Monumenta is a series of seven plate books published by the Society of Antiquaries of London between 1718 and 1906. The print volumes of Vetusta Monumenta contain 337 copper plate engravings that depict antiquities discovered across the British Isles. The digital edition of Vetusta Monumenta seeks to explore the methods members of the Society of Antiquaries used to select objects to include in these volumes. Using digital surrogates of the plates contained in the printed volumes has required developing metadata practices that reflect the breadth and depth of the plate books’ contents. Using the Visual Resource Association (VRA) metadata Import/Export panel facilitates the creation of granular and interoperable metadata, which supports the creation of metadata crosswalk and extraction tools in Scalar, the project’s digital publishing platform. Initially, creating granular metadata in Scalar required manually translating VRA Core metadata into Dublin Core and IPTC terms. Working closely with Scalar developers at the Alliance for Networking Visual Culture (ANVC) has made it possible to automate this process. Automating metadata processes has supported experimenting with annotation and visualization tools for both textual and graphical components of the digital edition. Over the past five years, working closely with developers at ANVC has helped to clarify and refine metadata workflows that support the production of linked data. Recognizing the dynamic role metadata can play in the organization and interaction of contents has facilitated ongoing discussions about methods for creating and sharing linked data. For instance, one significant outcome of creating and maintaining granular metadata is a dynamic relationship between scholarly commentaries and image files (and vice versa). Using metadata to describe the contents and structural characteristics of media files has supported augmenting media’s function in the digital edition. Currently, image and text annotation produce visualizations, which facilitates the development of innovative search and display interfaces that allow users to explore the digital edition’s contents based on geographic, thematic and subject related contents. This presentation will highlight the interdisciplinary nature of the digital edition, and will explore the contributions metadata can make to digital humanities research and work.

Presentation Type: 
Talk
Language: 
English
Presenter: 
Audio Size: 
21.6
Presentation Visual: