%oddex; %dtdmods; ]>
TEI ED W23: Outline of TEI P2. No source; created in m-r form. 22 Sep 92 : MSM, LB : agree on chapter inventory, sequence, IDs 21 Jul 92 : CMSMcQ finish mods from tech review May 92, revise 3-4 Jun 92 : CMSMcQ modify toc as per tech review May 92, retag as p2x file, rename fsd as feature *system* decl (proposal) 22 Feb 92 : CMSMcQ modify table of contents as agreed Jan92 31 Dec 91 : CMSMcQ split long lines so all lines < 80 chars 2 Dec 91 : CMSMcQ specify ID, filename, and initials of editor responsible for initial revision 24 Oct 91 : CMSMcQ further revisions in Bonn with particular attention to front and back matter 8 Oct 91 : LB sent back to Chicago 28 Sep 91 : LB,MSM merged with LB's notes, reformulated 4 Sep 91 : CMSMcQ made file
Outline of TEI P2 C. M. Sperberg-McQueen Lou Burnard TEI ED W23 &docdate Abstract

This document describes the content projected for document TEI P2, the second public draft of the TEI Guidelines. The overall structure of the guidelines is not expected to change dramatically between the preparation of TEI P2 and the final version of the guidelines to be submitted to the Advisory Board for endorsement at the completion of the current development cycle. Overall Types of Content

The work products of the first and second cycles must contain or be accompanied by: Tutorial introduction to markup concepts, computer representation of texts, SGML, and document processing (similar to chapter 2 of TEI P1, to ED W21, and to various other introductory documents, mostly by LB) Prose specification of the TEI encoding scheme arranged by topic (similar to chapters 3-8 of TEI P1, without some of the more discursive sections). This prose should be more abrupt than that of TEI P1 (more like that of the presentation of tags in ED W18 and ED W21). Prose commentary on the problems raised by tag sets for particular areas (similar to the discursive digressions in TEI P1) Alphabetically arranged reference section listing each tag, attribute and special attribute values alphabetically, with ample cross references Formal DTDs or DTD fragments for the TEI scheme Case studies, containing extended examples, with commentary, of the use of the TEI scheme, preferably on real data Overall Organization

We propose to organize these into documents as follows: TEI P2: The Guidelines Prose Specification

The prose specification portion of the Guidelines should contain: introduction, discussion of structure; status of scheme; notions of 'core', 'base', and 'topping' chapters on the TEI Core, with material relevant for all users chapters on alternate bases chapters on additional tag sets See the summary table of contents below for the specific topics to be covered, and their arrangement. Canonical Structure of Prose Presentation of Tags

Each section on a set of tags should, in order: define/describe the set of textual features covered in the section, in a paragraph or two give, in list form: the tags used to mark the features (as in 5.3.2 and many other sections of P1) following each tag, its non-global attributes if any (as in 5.3.9 et al. of P1) following each attribute description, any special values (as in 5.3.3 and 5.6.4 of P1, inter alia) if the attribute has a semi-closed domain One, two, or three examples of at most a paragraph in length. In rare cases, these may include an image of the relevant page of the copy text(s). One or more DTD fragments defining the tags and attributes. Reference Section

This section, modeled on the manuals for Formex and Majour, will have entries for each tag and attribute modeled on the specification in ED W5, plus information on what set of tags the tag fits into (core, prose base, text criticism, ...) and whether it is required, recommended, or optional. A short example should be provided for each tag; the same example may be reused for several tags. DTDs

These will be presented both (a) much as in the current appendix B, and (b) with alternation of prose commentary and DTD, roughly as in Donald Knuth's publications of Web, TeX and Metafont. It has proved feasible to make a version of Knuth's Web system to work with SGML documents and DTDs (see TEI ED W29, the documentation for the ODD system), and that system is used to generate both the DTD fragments and the full DTDs. Summary Table of Contents for TEI P2

The table of contents for TEI P2 will thus resemble the following; for each section, the list gives its identifier (the value of the id attribute in the div) and the initials of those with initial responsibility for drafting and first-pass editing. Where appropriate, the id, filename, etc. assigned by previous versions of this document are also given.

Filenames of chapters will consist of P2 followed by the chapter's id. Subsections will append characters or digits to the id and file name of their containing section: subsections 1, 2, and 3 of chapter 22 (p2ch) may have ids p2ch1, p2ch2, p2ch3. File names for subsections, like file names for chapters, consist of the string p2 followed by the section id. As was the case with TEI P1, each div1 or div2 should be in a distinct file. (The div1 chapters embed the files for their daughter div2s.) Front matter Part I: Introduction Part II: Core Tags and General Rules Part III: Base Tag Sets Part IV: Additional Tag Sets Part V: Auxiliary Document Types Part VI: Technical Topics Part VII: Alphabetical Reference List of Tags and Attributes Part VIII: Reference Material back matter

Moved or removed are: Tags for Analysis and Interpretation (was: id=p2.44, draft=DTL, ed=MSM) (split to SA, AI, FS) Applications of Analytic Tools (was: id=p2.45, draft=DTL, DGr, DR and DRoss, ed=LB) (to casebook) Hypermedia (TR3) (was: id=p2.42, file=P242, draft=SJD, ed=LB) (incorporated into SA) TEI U1 et seq.: Tutorials

Each tutorial should cover the following: Summary of basics of SGML, containing essentially the same material as chapter 3 of P1, but shorter, and with examples related to a specific discipline. Brief description of how SGML can be used, again with applications drawn from a specific discipline, and some account of existing software Brief summary of TEI philosophy, overall structure of the Guidelines, the TEI core tagset and any specialist tagset of particular importance to the disciplines A single worked example throughout each text Total length: not to exceed 30 pages. May be automatically generated, for the most part, from existing documents (e.g. EDW21, EDW24). It should be possible for a user to read this material alone and follow its advice, without referring to the rest of the book and without being too seriously misled.

At least one generic tutorial should also be available by June 1992. Other disciplines we hope to cover are: literature, history, linguistics and lexicography. Co-operation from specialists in these fields would obviously be very useful. TEI T1 et seq.: Case Books

These will resemble those in appendix C of P1, but should be accompanied by prose commentary. The samples prepared by the affiliated projects are good candidates for inclusion as extended examples, as are the transduction papers being produced by ML.