Changes from TEI P1 to TEI P3

This list gives a partial indication of the major changes from Versions 1 and 2 of these Guidelines (issued by the TEI as drafts under the document numbers TEI P1 and TEI P2, the latter released in chapters between March 1992 and the end of 1993) to the current text.

Chapter 1 : this chapter corresponds to chapter 1 of TEI P1 and TEI P2; it has been reorganized, revised, and expanded, and a new section explaining the notational conventions of this document has been added.

Chapter 2 : this is a slightly revised version of chapter 2 of TEI P1 and P2. Brief discussions of parameter entities and marked sections have been added, but no other changes of substance have been made.

Chapter 3 : this chapter was introduced in TEI P2; the lists of classes and important parameter entities have been updated, and some declarations have been reordered; no other changes have been made.

Chapter 4 : this chapter corresponds to some material in chapter 3 of TEI P1, but presents it in what is hoped to be a more accessible form. No substantive changes have been made since its publication as part of TEI P2.

Chapter 5 : this is a revised and much expanded version of chapter 4 of TEI P1. The overall structure of the TEI header has been retained, but most of the elements have been renamed to match a new set of naming conventions. The <encoding.declarations> element of TEI P1 has been split into the <encodingDesc> and <profileDesc> elements, the former concentrating on the process by which the electronic text has been encoded, the latter on the non-bibliographic characteristics of the text itself. A number of specialized declarations have been added to both these sections of the header, in order to allow the formal specification of important information about the text and its encoding.

Chapter 6 : this chapter corresponds to sections 5.3 to 5.6, portions of 5.7, and 5.8 of the first public draft of these Guidelines (TEI P1). Changes made to this material in this version include:

• The individual sections have been reordered and reorganized.
• Highlighting and quotation marks are treated together.
• The tags for names and dates have been revised, and a separate additional tag set has been provided for detailed analysis of names and dates (chapter 20 ).
• The tags for simple editorial interventions have been revised; the new set includes several complementary pairs of elements, so that the encoder is consistently given the choice of recording the original text, or an editorial modification of it, as data content, and the other as an optional attribute value.
• The tags for bibliographic references have been renamed (from <citn> to <bibl> , etc.) and a new form (<biblFull> ), corresponding to the structure of the TEI header, has been added.
• The treatment of canonical reference systems has been thoroughly revised and the discussion is now supplemented by discussions in chapter 5 , and chapter 32 .

Chapter 7 : this chapter corresponds to section 5.2 of TEI P1. Changes made to this material in this version include:

• The theoretical discussion of alternative methods of constructing a tag set for overall text structure has been suppressed.
• The tags for elements of a title-page have been renamed.
• The specialized tags for divisions of front matter and back matter (<foreword> , <acknowledgements> , <dedication> , <colophon> , etc.) have been deleted; like those of the text body, these elements may be tagged with generic <div> elements.
• In addition to numbered <div> elements, the current draft also allows for un-numbered generic <div> s.
• The treatment of collections and anthologies is explicitly discussed, building on section 7.2 of TEI P1, and the <group> element is introduced to deal with them.

Chapter 8 , chapter 9 , and chapter 10 : these correspond to the subparts of section 7.3 of TEI P1, but have been completely redesigned and rewritten from scratch.

Chapter 11 : this chapter first appeared in TEI P2; it has been revised here to match changes in the overall design of the Guidelines since its publication. Most importantly, this tag set now uses the default text-structure elements described in chapter 7 , and the methods for handling overlap and other time-specific information have been revised to make use of the techniques described in chapter 14 .

Chapter 12 : the tag set presented in this chapter is a complete revision of that described in section 7.4 of TEI P1, and the chapter itself was entirely rewritten from scratch.

Chapter 13 : this chapter was first published in December, 1993, as part of TEI P2. Since that publication, it has been revised slightly for the sake of consistency with the rest of the Guidelines and with the work of Technical Committee 37 of the International Organization for Standardization (ISO) on ISO DIS 12 200.

Chapter 14 : this chapter corresponds to sections 5.7 (``Links and Cross References'') and sections 6.2.3 through 6.2.5 (``Alignment of Multiple Analyses,'' etc.) of TEI P1. Changes made to this material in this version include:

• The <xref> element of P1 has been split into the two elements <ptr> and <xptr> , of which the former is used to point at IDs within the document and the latter for pointing outside the document or for pointing at passages without IDs in the current document.
• The elements <ref> and <xref> have been added to provide pointer elements which can accept character content, for cases in which the pointing phrase of the source text cannot be reconstructed algorithmically.
• The `extended pointer syntax' has been substantially revised and systematized; the syntax and semantics of extended pointers have been defined more precisely.
• The <unit> and <level> elements defined by P1 for implicit alignment of multiple levels of analysis have been dropped; in their stead, the revised feature structure elements should be used. These are defined in chapter 16 .
• The elements <alignment> , <al.map> , <al.ptr> , <al.list> , <al.range> , defined in P1 for explicit alignment of multiple texts or analyses, have been replaced by the <link> , <linkGrp> , <corresp> and <correspGrp> elements. Because the <link> and <corresp> elements can point at multiple targets, there is no need for special alignment pointers, alignment lists, or alignment range elements.
• The elements <link> and <correspond> may be used in connection with <xptr> to align elements in external entities or passages which do not bear SGML identifiers.

Chapter 15 : the bulk of this chapter is new, though some parts of its substance derive from chapter 6 of TEI P1. The global ana and inst attributes have been added, to simplify the notation for simple forms of alignment between text and analyses; the elements <span> and <interp> have been introduced, to simplify the specification of analyses which do not require the structural rigor of feature structures.

Chapter 16 : this chapter derives from sections 6.2.1 and 6.3 of TEI P1. In its broad outlines, the feature structure notation introduced there is retained. The most important changes include these:

• Some elements have been renamed.
• Feature names and feature structure names are now represented as attribute values, rather than as embedded subelements.
• The treatment of Boolean logic has been substantially changed.

Chapter 17 : this chapter was introduced in TEI P2; its wording has been revised slightly since then, but the tags described remain the same.

Chapter 18 : this chapter presents new material on the use of the core tags for editorial intervention, and on specialized problems in the transcription of primary source material, especially manuscripts.

Chapter 19 : this chapter is a substantial revision of section 5.10 (``Critical Apparatus'') of TEI P1. The major changes include the following:

• The single end-point attachment method of encoding critical apparatus has been dropped.
• A new method of apparatus encoding, the location-referenced method, has been introduced to simplify the transcription of existing critical editions.
• The problem of subvariation is treated more explicitly.
• The <witList> element has been introduced for the purpose of identifying all the witnesses whose readings are recorded in the apparatus.
• The treatment of detailed information about a particular reading in a particular witness (in the <witDetail> element) has been changed somewhat.

Chapter 20 : this chapter is new in this version.

Chapter 22 : this chapter replaces section 5.9 of TEI P1; it provides small but usable tag set for tables, and describes in much more detail the process of including graphical information (figures, illustrations, etc.) in TEI-encoded texts.

Chapter 23 : this chapter builds on section 7.2 of TEI P1, but its contents are largely new. The tag set described here provides much fuller methods for documenting text type, subject area, and demographic characteristics of speakers, listeners, authors, etc. associated with the texts of a corpus.

Chapter 24 : this chapter was introduced in TEI P2; it has been slightly revised since.

Chapter 25 : this chapter derives from the writing system declaration described in chapter 3 (``Characters and Character Sets'') of TEI P1. The structure of the WSD has been changed slightly, and the chapter now gives an explicit account of the semantics of specifying base character sets, entity sets, or WSDs, and of modifying them using the <exceptions> element.

Chapter 26 : this chapter is new in this version of these Guidelines.

Chapter 27 : this chapter first appeared in TEI P2; it has not changed substantially since.

Chapter 28 , chapter 29 , chapter 30 , chapter 31 , and chapter 32 : these chapters are all new in the current version of these Guidelines (though the mechanisms of modifying the TEI DTDs described in chapter 29 remain the same as those described in chapter 8 of TEI P1). The definition of conformance provided in this version of these Guidelines differs from that of TEI P1 primarily in making more explicit the nature of the requirement that extensions to the tag set be documented, in specifying the nature of the DTD modifications allowed in TEI-conformant documents, and in completely divorcing the issue of TEI-conformance from that of the character sets used in the document.

The alphabetical reference list of classes, entities, and elements was introduced in TEI P2; in this version, slightly fuller information is given. For element classes, lists of members are given which include members of all subclasses, and the declarations of the a-dot and m-dot parameter entities for the class are reproduced. The files in which entities and elements are declared are also given.

Chapter 39 : this chapter appeared in TEI P2 and has not been revised for this version of these Guidelines.