TEI and Microsoft: a marriage made in...: Using proprietary tools for producing standardly encoded editions

1. XML and editorial interventions It is generally assumed that scholarly annotated digital texts, be they text-critical editions or linguistically annotated corpora, should be stored in XML to ensure longevity as well as platform and software independence. Furthermore, the TEI Guidelines (SperbergMcQueen and Burnard, 2002) are the de-facto standard for constructing the XML schemas for most such texts. But while XML is well-suited for machine processing, it is less than ideal for authorial or editorial interventions into the text, esp. when used with the complex TEI-derived schemas. It is of course possible to edit XML documents directly in a plain text editor, or better yet in specialised XML editors that support on-the-fly validation against a schema and schema-dependent drop-down menus. But, depending on the text type and required manual interventions, these generic editors are too clumsy for extensive work, and do not enable complex constraints on the allowed content and changes in the annotations. An additional problem with using XML editors for editorial work is the fact that many humanities scholars or students who are most likely to be doing this work have no knowledge of XML or TEI or any experience in editing it. This problem becomes all the more relevant in collaborative projects in which, say, a large number of students are hired to annotate a certain text or text collection. The effort required to first teach them how to use an XML editor and the underlying concepts might be prohibitive, and saving time necessary to perform each editorial intervention is essential.