ODD

ODD stands for "One Document Does it all". It is a TEI XML-conformant specification format that allows one to customize TEI P5 in a literate programming fashion. It uses elements from the new Documentation Elements module.

Description
The TEI Guidelines, its DTD, and its schema fragments, are all produced from a single XML resource containing:


 * 1) Descriptive prose (lots of it)
 * 2) Examples of usage (plenty)
 * 3) Formal declarations for components of the TEI Abstract Model:
 * 4) elements and attributes
 * 5) modules
 * 6) classes and macros

We call this resource an ODD (One Document Does it all), although the master source is instantiated as a gazillion XML mini-documents.

A system of XSLT stylesheets called Roma has been created for the purpose of easy manipulation of ODD files.

Example
The TEI scheme can only be used by customizing it. Customizations are also expressed in the ODD language. For example:

 This is TEI Lite with simplified heads         

produces a schema a bit like TEI Lite, with a slight change : the element &lt;head> now can contain only text nodes.

Uses besides the TEI Guidelines and various customizations

 * Internationalization Tag Set (ITS)
 * various standards proposal designed within ISO committee TC 37 have been totally or partially written in TEI/ODD: MLIF, MAF, ISO 16642 rev., ISOTimeML

Future plans: "Pure ODD"
Over the years ODD has evolved in response to the needs of the TEI community, both as the core specification language for the TEI guidelines, and also as a generic customization language. Since 2004, much effort has been put into making it less of a hybrid language, combining its own vocabulary with RelaxNG fragments and Schematron rules. Versions of TEI P5 later than 2.5 contain elements which can be used to express content models directly in the TEI ODD language, rather than in RELAXNG. This work is ongoing in the TEI Technical Council, the goal being to make ODD an entirely self-contained generic specification language, independent of other existing grammars, and hence known as Pure ODD. See further "Pure ODD". There is a mailing list for discussing Pure ODD called tei-odds.

The proposals for Pure ODD were finally included in the Guidelines during January 2016, for inclusion as part of the 3.0.0 release

Now that the project has been completed, make sure to go on to read How to Update your ODD

Documentation

 * Getting Started with P5 ODDs
 * How to Make an ODD Automagically
 * ODD chaining for Beginners
 * How to Update your ODD (from a pre 3.0 version)
 * A fragment of W3C's Best Practices for XML Internationalization concerning ODD customization for the Internationalization Tag Set (ITS).
 * Mapping ODD processing

Presentations

 * A talk about the ODD system, given on 13 Feb 2007 by Lou Burnard and Sebastian Rahtz at the OUCS Encoding Digital Texts workshop
 * A PowerPoint presentation on TEI/ODD by Laurent Romary; OUCS, September 2006
 * Literate serialization of linguistic metamodels by Piotr Bański, Balisage-2011

Articles

 * Resolving the Durand Conundrum, by Lou Burnard
 * Odd Customizations, by Syd Bauman and Julia Flanders.
 * RelaxNG with Son of ODD, by Lou Burnard and Sebastian Rahtz.
 * Freedom to Constrain by Syd Bauman