The TEI Council is in charge of overseeing the technical development of the TEI Guidelines. As in the past, during the current year, all energy was devoted to develop P5, slated to become the next major revision of the Guidelines. While in the past development took place mainly in workgroups charged by the Council, this year the Council took on more and more tasks by itself and did not charge any new workgroups this year. The Council held one face to face meeting with all its (available) members and another meeting, dubbed the "class meeting" of interested members to tackle the long-due overhaul of the structure of the TEI class system. The Council also conducted its regular business through telephone conferences, a mailing list and through an IRC channel.
The highlights of the TEI development work in the period since the previous Members meeting can be summarized as follows:
Information about the status of P5, how to use it and much more is available at the P5 section of the TEI website.
As indicated above, the work of the Council focussed this year on ironing out the many messy details that will hopefully provide a stable infrastructure for the future development of the TEI schema. This whole process can be likened to a restoration project of a venerable piece of architecture: When I grew up in the small southwest German town of Tübingen, conservation and re-creation of the medieval town centre was a major project there. This resulted in many a half-timbered house being completely redone from within without actually taking down the structure, but resulting in a totally new interior, with the facade and most exterior face-lifted, but essentially the same as before. Something similar is happening at the moment with P5, which will be a similarily re-furbished version of P4, with all the amenities a modern inhabitant does expect, but without sacrifying the time-proven and time-honored facade.
For the purpose of this report, I will focus on three areas that witnessed particular intensive development this year:
One of the major goals for TEI P5 has become the systematic and thorough removal of plain CDATA attribute values. In the new P5 world, attribute values will be of a defined datatype wherever possible, while many of P4 CDATA attribute values that used to hold textual values (for example @sic on <corr>) have been moved to element nodes of their own. This became necessary to overcome limitations inherent to the XML recommendation with respect to indicating a language for attribute nodes and markup contained in attribute nodes.
It became thus necessary to review all attributes on all elements defined within the TEI scheme and decide what datatype to use for them. The document used as a base for these decisions, ED W 90 is available, as well as the table of attribute values. TEI members are encouraged to look at these documents and give feedback to the Council or Editors.
The ‘ODD system’ refers to the new chapter in P5 Documentation Elements, which describes a module of the TEI schema that can be used to document and declare XML elements and element classes. It is also used in the TEI Guidelines to describe the TEI schema itself and can, with the help of an ‘ODD processor’ (which is known under the name ‘roma’, available both as a web service and as a shell script for UNIX-like environments from Sourceforge) be used to generate machine readable schemas in DTD, RNG or XSD fomat, as well as the accompanying documentation.
Work of the Council in this area concentrated on the generalization and abstraction of the whole system, as well as its processor, to enable it to serve as an easy and convenient tool to describe TEI customizations and to make sure that it will be extendable to support future multilingual versions of such customizations.
Work on the ODD system made it clear that the whole TEI class system, which was build to support the old ‘Chicago Pizza Model’, needed a complete overhaul. Members of the Council devoted a whole meeting to this question and considered both general rules for creating and naming classes as well as implications of this for the elements of selected modules. The work done so far is reflected in the most recent release, but it has to be considered experimental and further experience is required here.
Apart from the major issues metioned above, many issues of smaller scale have been tackled as well. This includes a proposal of a group of scholars working with Sanskrit texts, which will allow multiple conflicting analytic descriptions of a text, work on aligning the various elements and recommendations for bibliographic records to modern practice and an attempt to map metadata standards to the elements in the TEIHeader systematically (this has been taken care of by Natasha Smith and John Walsh).
The Council also worked with vendors of XML editors willing to include TEI schemas and stylesheets with the products and developed a recommended directory tree structure for the TEI deliverables, the canonical location for this is http://www.tei-c.org/release/ and is committed to the continued maintenance of its bundle tei-emacs, available from the TEI Software page.
Progress on this workgroup has been slow, and there remain many things yet to do. The work is carried on by email list and conference call. Energetic volunteers are encouraged to contact the chair if they want to participate.
A list of documents produced by this workgroup is available in the SO section of the Activities area on the TEI website.
This is a joint activity of the TEI together with the ISO TC37/SC4 workgroup on Feature Structures. The workgroup met in Warszaw Aug. 2005 , where it was reported that the results of their work is now officially a draft international standard, and that it received almost unanimous support. Much of the credit goes to the chair of the ISO workgroup, Kiyong Lee, who shepherded this chapter through the sometimes daunting challenges of ISO procedure. There was also agreement that there should be a new work item for ISO to do a revision in a similar way for the feature system declaration. This will be part 2 of ISO 24610. This might start again with the TEI chapter as a basis, but other ontology definition languages, automatic validation tools, etc. will also be investigated. Lou Burnard will continue to be the TEI liaison.
A list of documents produced by this workgroup is available in the FS section of the Activities area on the TEI website.
Sebastian Rahtz has been invited to jon the W3C workgroup on internationalization of tag sets, which is chaired by Richard Ishida. The group held a meeting in Nice in September 2005. For more information about this effort, please see the website of the WG.