Semi-structured data processing: implementation hypotheses and use cases taken from Old English texts

Abstract

Although EVT has slowly evolved into a versatile and powerful tool for publishing historical sources and literary texts on the basis of the XML/TEI format, none of the modifications - and periodic rewrites - of its code base has significantly changed the way in which edition data is handled. Even today, in fact, the main goal is to visualise and navigate such data, no doubt in a sophisticated way (support for multiple edition levels, support for named entities management, text-image linking), but in any case limited to what we might call “core functionalities” of a DSE. Texts encoded in XML/TEI, on the other hand, are a potential treasure trove of information just waiting to be interrogated and made available to the user. In this article I identify three use cases – the processing of special characters encoded by means of the <charDecl> and <g> elements; the management of named entities, realia and other interesting elements of the text; the use of ontologies within an XML/TEI document – related to literary texts in Old English in order to propose a TEI encoding and a subsequent processing able to put the user in a position to receive answers to complex, transversal queries involving cross-links between different types of elements.