TEI Text Documentation Committee
           Summary of L. Burnard's Proposal for
                  Revision of TEI Header
 
 
1. Normalize naming conventions, notably:
 
     use .desc suffix for three major divisions of
     header, thus file.desc, encoding.desc, and
     revision(s).desc
 
     use .group suffix for major groups (not .statement,
     .declarations., etc.)
 
     use .decl consistently for declarations of encoding
     practice to which reference may be made from
     elsewhere (e.g.(sampling.decl), not (sampling.def),
     (editorial.decl) or (editorialPrinciples.decl) not
     (editorial.principles)
 
     use .type consistently for references to declarations
     made with .decl elements (as in the current
     (encoding.type) or (sampling.type)
 
2. Shorten various names {LB will provide specific list}
 
3. Revise encoding declarations (pp. 66-67 in TEI P1) in
light of TR6 and other recommendations
 
4. Provide separate tags for some notes of particular
importance, e.g.
 
     - (availability) for availability information
     - (identification) (or another name) for ISBN, ISSN,
     other identifying numbers
 
5. Where applicable, state explicitly what MARC fields a
given TEI header tag corresponds to.
 
 
 
              TEI Text Documentation Committee
 
        Proposal for Changes in Tag Names (L. Burnard)
 
This note sets out some specific proposals for change in the tag
names used within the full TEI header. Most of the changes are
motivated by a desire to find shorter names with a similar
expressive power to those currently proposed, and also to make
clearer the distinction between grouping and terminal tags. I
apologise in advance for not having had time to think through all
of the proposals: they are intended for discussion only.
 
1. The suffix ".desc" should be used for the high level grouping
tags.
This implies the following changes:
 
file.description -> file.desc
encoding.declarations -> coding.desc
revision.history -> revision.desc
 
2. The suffix ".group" should be used for groupings within each
heading rather than the current mixture of ".statement",
".declarations" etc. The suffix should only be used for tags
whose content is more than simply %soup; tags bearing it should
where possible be fully minimisable e.g. it should not be
necessary to supply both title.group and title if there is only a
title to be given.
 
This implies the following changes
 
title.statement -> title.group
edition.statement -> edition.group
extent.statement -> extent
publication.statement -> publication.group
 
series.statement -> series.group
notes.statement -> notes.group
 
 -- this one is an oddity: the dtd allows for %soup but states in
    a comment that it should contain only NOTEs --
 
source.description -> source.group
statement.of.responsibility -> resp.group
text.category.defs -> texttype.group
 
3. The following shorter names are suggested for some specific
tags:
 
principal.researcher -> principal
created.by -> creator
creation.date -> date
 
 -- the SGML context should be used to determine which date is
intended for all the different parts of the publication.group
element --
 
4. The suffix "decl." should be used for all tags used to provide
declarations that are invoked (by ID/IDREF) elsewhere in the
text. There are several such tags proposed by other work groups
not yet incorporated into P1; these are not specified in the
following list.
 
sampling.def -> sampling.decl
editorial.principles -> editorial.decl
reference.system -> refs.decl
category.definition -> texttype.decl
 
5. The suffix ".type" should be used for all tags used to invoke
tags in the group above. These tags should all use the same
attribute (suggested name 'target') to supply an IDREF pointing
to the appropriate tag from the group above. This requires only
the following changes:
 
ref.type -> refs.type
text.category -> category.type
 
6. Much of the discussion on pp. 66-67 of P1 is rendered obsolete
by the new proposals from WGs TR6, AI2, and TR12. Unfortunately I
have not had time to digest all of their new suggestions: I hope
you will find time in the meeting to make a start on them!
 
7. It would be good to introduce some more specific tags into the
section on 'general notes' proposed on p. 64 of the Guidelines.
Particular candidates include the following:
availability (MARC tag 506)
identification (MARC tag 510) -- we give ISSN as a component of
serial only, but surely it should be generalisable to any
bibliographic entity. Should probably have a 'type' and 'n'
attribute?
 
8. A minor suggestion but I think a helpful one: would it not be
a good idea to provide in the discussion of each group of tags in
this section a statement of the corresponding MARC tag where is
clear? My initial (very incomplete) mapping follows:
 
author -> 100
title.statement -> 245
edition.statement -> 250
extent.statement -> 256
source.description -> 537? 830?
encoding.declarations -> 538 (?)
statement.of.responsibility -> 700
keywords -> 690