TEI : Text Encoding Initiative

1
TEI Text Encoding Initiative

• Computing in the Humanities

July 11, 2003 John A. Mess
2
Overview

• Origins of Text Encoding Initiative
• Features of Encoded Text
• Tagging of Text
• Examples

Adapted in part from Lou Burnards
Workshop http//www.tei-c.org/Talks/ESS2001/elsnet
-4.htm Project Gutenbergs TEI Guide for HTML
Writers Guild http//gutenberg.hwg.org/teidtds.htm
l
3
Text Encoding Initiative

• 1988, the date of the Poughkeepsie Conference
• 1999, when the process of setting up the TEI
  Consortium began
• Originally, a research project within the
  humanities
• Sponsored by leading professional associations
• Major influences of
• digital libraries and text collections
• language corpora
• scholarly datasets
• International consortium established June 1999
  (see http//www.tei-c.org/)

4
Goals of the TEI

• better interchange and integration of scholarly
  data
• support for all texts, in all languages, from all
  periods
• guidance for the perplexed
• what to encode
• user-driven codification of existing best
  practice
• assistance for the specialist how to encode
• loose framework into which unpredictable
  extensions can be fitted
• These apparently incompatible goals result in a
  highly flexible, modular, environment for DTD
  customization.

5
TEI Deliverables

• A set of recommendations for text encoding,
  covering both generic text structures and some
  highly specific areas based on (but not limited
  by) existing practice
• A very large collection of element
  definitionscombined into a very loose document
  type declaration
• A mechanism for creating multiple views (DTDs) of
  the foregoing

One such view and associated tutorial TEI Lite
(http//www.tei-c.org/TEI/Lite/) for the full
picture see http//www.tei-c.org/TEI/Guidelines/
6
Legacy of the TEI

• Way of looking at what text really is a
  codification of current scholarly practice
• A set of shared assumptions and priorities about
  the digital agenda
• focus on content and function (rather
  thanpresentation)
• identify generic solutions (rather
  thanapplication-specific ones)

7
Designing a DTD for the TEI

• How can a single mark-up scheme handle a large
  variety of requirements ?
• all texts are alike
• every text is different
• learn from the database designers
• one construct, many views
• each view a selection from the whole

8
The Chicago Pizza Model

• A useful metaphor for expressing modularity now
  implemented athttp//www.hcu.ox.ac.uk/TEI/pizza.h
  tml lt!ENTITY base "deepDishthinCruststuffed"
  gt lt!ENTITY topping "pepperonimushrooms
  sausageanchovies... " gt lt!ELEMENT pizza (
  base, tomatoSauce, cheese, (topping)) gt

9
To build a TEI pizza, take...

• The core tagsets, the base of your choice, the
  toppings of your choice, (optionally) a reference
  to your extensions
• lt? xml version1.0 ?gt
• lt!DOCTYPE tei.2 SYSTEM tei2.dtdgt
• lttei.2gt
• ltteiHeadergt .... lt/teiHeadergt
• lttextgt .... lt/textgt
• lt/tei.2gt

10
The core tagsets

• Detailed metadata provision the TEI Header tags
  for a large set of common textual requirements
• paragraphs
• highlighted phrases
• names, dates, number, abbreviations...
• editorial tags
• notes, cross-references, bibliography
• verse and drama

11
The base tagsets define

• basic high-level structure of document one must
  be chosen from
• prose, verse, or drama
• transcribed speech
• dictionaries and terminology
• or combine two or more using either of the
  general base (anything anywhere)
• the mixed base (homogenous divisions)

12
TEI additional tagsets

• sets of elements for specialised application
  areas can be mixed and matched freely
• currently provided
• linking and alignment
• analysis feature structures certainty
• physical transcription
• textual criticism, names and dates
• graphs and trees figures and tables
• language corpora....
• in preparation...
• manuscript description

13
The Lampeter corpus in the Oxford Text Archive

• Fairly typical requirements for language corpora
• light presentational tagging
• structural markup for access
• demographic information about text production
• small number of tags to ease data capture and
  validation
• Implementation tagsets
• prose base, and
• tags from four additional sets some extensions,
  many exclusions

14
Issues with the TEI

• Unmodified TEI offers authors too many choices
• four different types of bibliographic citation
• three (or four) different tags for proper names
• an indigestably rich choice of text editing tags
• At the same time, unmodified TEI lacks
• detailed table model
• detailed tags for mathematical and other formulae
  
• front matter for modern publications
• tags for multimedia objects
• All this can be addressed by TEI customization

15
Why bother?

• The TEI is a well-known reference point
• Using the TEI enables
• sharing of data and resources
• shared modular software development
• lower learning curve and reduced training costs
• The TEI is stable, rigorous, and well-documented
• The TEI is also flexible, customizable, and
  extensible in documented ways
• The architectural approach offers the best
  compromise for practical work.

16
Using the TEI for authoring

• A DTD for authoring should be prescriptive rather
  than descriptive closely tied to current
  authoring practice very easy to use
• This suggests that we need content-full tagging
  only the tags we need and all the tags we need
• For details of version 4, see
• http//www.tei-c.org/P4X/index.html
• http//etext.lib.virginia.edu/tei/uvatei.html

17
Graphical Layout
18
The overall structure of a unitary text

• ltTEI.2gt
• ltteiHeadergt lt!-- ... --gt lt/teiHeadergt
• lttextgt
• ltfrontgt
• lt!-- front matter of copy text goes here. --gt
• lt/frontgt
• ltbodygt
• lt!-- body of text goes here. --gt
• lt/bodygt
• ltbackgt
• lt!-- back matter of text, if any, here. --gt
• lt/backgt
• lt/textgt
• lt/TEI.2gt

19
Core Structural Elements

• lt!ELEMENT TEI.2
• (teiHeader, text) gt
• lt!ELEMENT teiHeader
• (fileDesc, encodingDesc,
  profileDesc, revisionDesc?) gt
• lt!ELEMENT text
• ((index interp interpGrp lb
  milestone pb gap
• anchor), (front, (index interp
  interpGrp lb milestone
• pb gap anchor))?, (body
  group), (index interp
• interpGrp lb milestone pb
  gap anchor), (back,
• (index interp interpGrp lb
  milestone pb gap
• anchor))?) gt
• lt!ELEMENT group
• ((argument byline docAuthor
  docDate epigraph head
• opener salute signed index
  interp interpGrp
• lb milestone pb gap
  anchor), (text group),

20
TEI Header structure

• ltteiHeadergt ltfileDescgt
  ltencodingDescgt ltprofileDescgt
  ltrevisionDescgt
• lt/teiHeadergt

21
The File Description ltfileDescgt

• Mandatory
• Supplies full description of the electronic file
  itself, and its source/s
• Must specify at least a title, a publication
  statement, and a source
• Use of authority control is advisable but not
  required

22
The File Description

• ltfileDescgt
• lttitleStmtgt
• lteditionStmtgt 250
• ltpublicationStmtgt
• ltextentgt 300
• ltsourceDescgt
• ltnotesStmtgt 786
• lt/fileDescgt

23
The source description

• May contain common TEI bibliographic
  elementselements ltbiblgt, ltbiblStructgt,
• or a nested file description ltbiblFullgt
• or a list ltlistBiblgt
• or a prose description
• or specialised elements for transcribed speech
• Or (for the born-digital document) simply the
  text Original

24
Crosswalks
lttitle typemaingt DC.title.main 246
ltauthorgt DC.creator.name 100
ltpublicationStmtgt DC.publisher.name 260
ltsourceDescgt DC.source 500,534
ltclassDeclgt DC.subject.schema 6xx
25
The Body of a Text
26
Front Matter

• lttextgt
• ltfrontgt
• lttitlePagegt
• ltdocTitlegt
• lttitlePartgtRIDERS OF THE PURPLE
  SAGElt/titlePartgt
• lt/docTitlegt
• ltdocAuthorgt
• ZANE GREY
• lt/docAuthorgt
• lt/titlePagegt
• lt/frontgt
• ...rest of book content here...
• lt/textgt

27
Within a Text

• ltdiv1 type"chapter"gt
• lthead n"1"gtCHAPTER I.lt/headgt
• lthead n"chaptitle"gtLASSITERlt/headgt
• ltpgt
• A sharp clip-crop of iron-shod hoofs deadened
  and died away, and
• clouds of yellow dust drifted from under the
  cottonwoods out over
• the sage.
• lt/pgt
• ...
• lt/div1gt

28
Multi-Part Books

• lttextgt
• ltfrontgt
• ...book front content here...
• lt/frontgt
• ltgroupgt
• lttextgt
• ltfrontgt ...part 1 front content
  here...lt/frontgt
• ltbodygt ...part 1 body content
  here...lt/bodygt
• ltbackgt ...part 1 back content (if
  any) here...lt/backgt
• lt/textgt
• lttextgt ...content of part 2...lt/textgt
• lttextgt ...content of part 3...lt/textgt
• ...etc...
• lt/groupgt
• ltbackgt
• ...book back content(if any) here...
• lt/backgt

29
Drama Markup

• ltstage type"enterance"gtEnter HELENAlt/stagegt
• ltspgt
• ltspeaker who"Hermia"gtHERMIAlt/speakergt
• ltlgtGod speed fair Helena! whither away?lt/lgt
• lt/spgt
• ltspgt
• ltspeaker who"Helena"gtHELENAlt/speakergt
• ltlgtCall you me fair? that fair again
  unsay.lt/lgt
• ltlgtDemetrius loves your fair O happy
  fair!lt/lgt
• ltlgtYour eyes are lode-stars and your
  tongue's sweet airlt/lgt
• ltlgtMore tuneable than lark to shepherd's
  ear,lt/lgt
• ltlgtWhen wheat is green, when hawthorn buds
  appear.lt/lgt
• ltlgtSickness is catching O, were favour
  so,lt/lgt
• ltlgtYours would I catch, fair Hermia, ere I
  golt/lgt

30
Examples from Lady in Boomtown

• ltpgtThe statistics on ltrs type"place"
  key"GOLD1"gtGoldfieldlt/rsgt are from an article by
  Charles F. Spillman
• that appeared in the ltrs type"place"
  key"NEVA1"gtNevadalt/rsgt News Letter of January 1,
  
• ltdate value"1916"gt1916lt/rsgt, a copy of which I
  had preserved. His figures corresponded to my own
  memory
• sufficiently for me to accept their accuracy.lt/pgt
• ltpgtDuring ltrs type"person" key"TASK1"gtTasker
  Oddieslt/rsgt lifetime we talked for many happy
  hours
• about the old days, and much I have written came
  out of those conversations.
• He also furnished me with campaign literature
  pertaining to
• himself and ltrs type"person" key"KEYP1"gtKey
  Pittmanlt/rsgt, together with magazine clippings
  about the
• battleship Nevada. All of these were returned to
  ltrs type"person" key"TASK1"gtSenator Oddielt/rsgt,
  so I
• have no record of the published sources from
  which they were clipped.
• Most of the gossip came to me through my brother
  and my brother-in-
• law, who joined us when ltrs type"place"
  key"TONO1"gtTonopahlt/rsgt began to boom. Whenever
  either
• one of them picked up ltqgtoff the streetlt/qgt an
  exciting bit of information,
• they shared it with me and I added it to my
  notes.lt/pgt