XML technologies in language documentation workflows

Abstract

More and more programs use XML formats for internal data storage, not only for interchange. This includes both general-purpose tools like MS Office and OpenOffice/LibreOffice and specialized linguistic software such as ELAN, EXMARaLDA, FLEx, Speech Analyzer, Arbil, WeSay, SayMore and so on. Thus more and more linguistic data are being created in XML, not just convertible to XML. Although not ideal (verbosity, high processing time), XML formats have a number of benefits to boost workflow efficiency. Importantly, XML documents can be processed with XSL transforms to get new data, remaining still in the realm of XML (the XSL transforms themselves are also XML and can be transformed by other XSL...), displayed as HTML or published into PDF. Finally, there are now mature free native-XML databases like eXist-db and BaseX which offer the full cycle of operations in one application with browser-based interface: store existing documents, browse and query data, create and edit data online, apply XSLT to publish. I will illustrate this with examples of transformations we used in language documentation workflow to convert interlinear texts in Archi (East Caucasian) between various formats including OpenOffice and FLEx. A connected issue which will be addressed is the need for an interchange standard format for interlinear texts.