Linking and Querying Ancient Texts: A multi-database case study with epigraphic corpora
Tobias Blanke, Mark Hedges, & Shrija Rajbhandari (King's College London)
Digital Classicist/ICS Work in Progress Seminar, Summer 2009
Friday 19th June at 16:30, in room STB3/6, Senate House, Malet Street, London WC1E 7HU
Databases and corpora of marked-up texts form a significant proportion of the outputs of digital archaeology. Although the development of standards for the representation of such information creates new possibilities for interoperability, significant problems remain. Legacy data exists in diverse and often obsolete formats, and even when standards are applied, the sheer variety of cultural data and research means that there remains a great deal of ‘fuzziness’. Users must be trained in the correct application of standards, entailing significant outlay of time and money. Moreover, there is inevitably diversity of representation when information is gathered together from different projects and for different purposes, and thus there will always be a need to integrate this diversity while preserving the integrity of the data itself.
The LaQuAT (Linking and Querying Ancient Texts) project investigated technologies for providing integrated SQL-based views of diverse data resources related to classical archaeology, specifically containing epigraphic and papyrological material. These resources are quite heterogeneous in terms of standards and structure, comprising two relational databases with different schemas, and an XML-based corpus; they are hosted by different institutions in different countries, and are the outputs of divergent research communities. Nevertheless, the datasets overlapped geographically, chronologically and prosopographically. Such issues are typical of archaeological databases; to address them the project explored the applicability of ‘grid computing’, and in particular the OGSA-DAI software, for providing integrated views of such diversity.
The seminar will be followed by wine and refreshments.