[note1]
Concerning the elements included from the additional TEI tagsets:
all elements from the
Linking, figures, transcr, and
textcrit additional tagsets were included. Of the Analysis
additional tagset, only <interp> and <interpGrp> were included. Of the
names.dates additional tagset, only <placeName>, <settlement>,
<region>, <country>, <bloc>, and <distance> were included.
[note2]
This section will
discuss the specific extensions of the TEI header that are
incorporated in the DALF header. This means that some
familiarity with the TEI header is assumed from the
reader. We considered it no use to repeat the excellent and
comprehensive documentation for the TEI header, the interested
(or desperate) reader can find at http://www.tei-c.org/P4X/HD.html. As
for the standard TEI elements occurring within specific DALF
header elements, a concise explanation is provided, as well as a
reference to the full TEI documentation of the elements concerned.
[note3]
A full
technical discussion can be found in Arjan Loeffers answer on
28 October 1997 to Andreas Nolda who formulated nearly the same
problem we encountered. These messages can be found in the TEI
List archive.
[note4]
This distribution
is based on that of the original TEI <figure> element. Note that there is a
slight discrepancy between the TEI documentation of this element (see
http://www.tei-c.org/P4X/ref-FIGURE.html) and its declaration in the DTD files. The
documentation states that <figure> is a member of only 2 element classes,
namely inter and tpParts; whereas the TEI
teiclas2.ent file declares <figure> as member of a third element class:
common. Since in practice the distribution of <figure>
clearly suggested the latter evidence, the DALF scheme also defines all elements
whose distribution is modelled after that of the original TEI <figure>
element as members of the common, inter, and
tpParts element classes.
[note5]
None of these options provides 100% satisfactory solutions,
however. Encoding multiple hierarchies in XML has become a generally
acknowledged problem for which a number of other solutions
(independent of TEI) are proposed, varying from degrees of
‘virtual markup’ (e.g. standoff-markup, Bottom-Up Virtual
Hierarchies) to abandoning XML for another encoding
scheme (MECS, TexMECS, Just-In-Time Trees). Up to now, this has
not lead to ready-to-use, generally accepted ways of dealing with
overlapping hierarchies in XML. Therefore, we stuck to the
proposals within the TEI paradigm, which provides after all the
best guarantee for future standard-compliance.
[note6]
In particular, the "suggestions for the forthcoming/ongoing (?) TEI revision"
made by Hilde Bøe, Ellen Nessheim and Stine Brenna Taugbø
for the Henrik Ibsen's Writings, posted on 4 December 2001. These included
(a.o.) the redefenition of <add>, <figure>, and <note>
as global elements. Also in his answer of 28 July 2000 to a question
regarding the impossibility to use <note> inside <salute>,
Syd Bauman describes this as "an error with TEI including the TEI-Lite view", and
suggests the option to extend the TEI DTD.
[note7]
See note 3 for a discussion on the distribution of the TEI <figure> element.
2005-05-04Edward Vanhoutte & Ron Van den Branden.
© Copyright CTB 2004
