The second half of section 2.7 (#HD8) describes divergences between the TEI header and national cataloguing codes such as AACR2. (It used to talk about only AACR2, but I've fixed that: http://tei.svn.sourceforge.net/viewvc/tei/trunk/P5/Source/Guidelines/en/HD-Header.xml?r1=11539&r2=11540 .)
This section grew out of suggested revisions a couple of us presented to Council in April 2010. After discussion, Council agreed to have Lou adopt the uncontroversial parts and have the ad-hoc committee bring back further reconsidered revisions. Lou did so, including a list of bulleted items that desribe differences in ontology. The committee has been working on this piecemeal, focusing on first straightening out use of <biblScope>. Now that that's done, I'm looking for things that we still haven't dealt with.
The proposal contained one more list item, which is commented out in the ODD source:
<!--item>The components of the TEI file
description do not prescribe that data be transcribed according to
AACR2 rules.</item--><!-- removed as already stated above-->
I do not feel that what I intended to say was adequately expressed above. So I would like to add the following, now as the first bullet in the list:
• The TEI Guidelines do not prescribe that text be transcribed from the chief source of information with normalized capitalization and punctuation as in a national cataloguing code.
I feel that a user should understand that the TEI doesn't tell you how to transribe any bit of text from the source that you put in a particular element in the header. So even if you used a TEI header to MARC crosswalk like the one found in the Best Practices for TEI in Libraries ( http://purl.oclc.org/NET/teiinlibraries ), or even if you use Thutmose II ( http://wiki.tei-c.org/index.php/Thutmose ), your resulting record won't look like other catalog records.