Scope
The first objective of this SIG is to study and to document the usability of the
existing TEI biblStruct element for the representation of
bibliographies, whether these are organized as academic reference lists citing
published material, or, in the case of humanities scholarship, references to
primary sources. Secondly, the SIG aims to identify a set of core encoding
guidelines to be used by scholars both in preparing such bibliographies and for
use in the relevant sections of the TEI header (e.g. sourceDesc) of
their TEI encoded sources. Thirdly, the SIG will develop mappings to and from
other bibliographic formats widely used in the research community such as MODS,
Refer, Bibtex… The SIG will not address the issue of citation styles and other
presentational aspects.
Context
The P5 edition of the TEI guidelines, after a lengthy debate on whether there was
need for another element (biblItem) for bibliographic descriptions, has
extended the expressive power of biblStruct most noticeably with the
relatedItem element, inspired from MODS. It has also defined a
number of element model classes. In its current state biblStruct seem
to offer the richest method of describing scientific works: independent works,
relation to issues and series, multiple editions etc.
Program of work
The activity of the SIG will essentially be example based. This both complements
the TEI guidelines, and is of direct benefit to users with basic needs and
questions concerning the use of biblStruct.
Gathering samples
The group will gather a collection of reference samples representing typical
cases in encoding scientific works or primary/historical sources. These
exempla will be manually encoded and documented onlineCf.
prior similar activitues carried out in German under
http://colab.mpdl.mpg.de/mediawiki/TEI_Bibliographic_Information.
Some of the important issues are the encoding of names (authors and/or
organisations, especially the separate encoding of for- and surnames,
pseudonyms etc.), the encoding of references to standardized name files like
the LoC Name Authority File or PND PND =
Personennamendatei (name authority file) is the german reference file
for personal names., the values for biblScope/@type (cf.
biblScope type=”pp” vs. biblScope type=”pages”) and
the encoding of ‘divided’ publications (eg. an journal-article which is
printed on p. 1–7, 20–26 and refers to figure 1–5 at the end of the
containing journal). Also, there could be a standardization of
bibliographical references and citations in texts (maybe by embedding
CiteSchemaCf. http://xbiblio.sourceforge.net/cite/
for a description of citeSchema. as a namespace).
Documenting formats – mappings
Sample exempla will also systematically be represented in other formats
widely used in scientific works (BibTex, MODS, Endnote/refer); this in turn
should lead to a systematic documentation of the possible mappings between
these.
Converters
Converters between biblStruct elements and other commonly used
formats will be developed using XSLT stylesheets.
Converters to citation-styles of academic journals etc. may be realized by a
XSLT stylesheet transforming biblStruct and using the Citation
Style Language (CSL).Cf.
http://xbiblio.sourceforge.net/csl/.See also
RefDB, http://refdb.sourceforge.net, for SGML, XML, bibtex, and
RTF
Related work
The SIG members will monitor and (where appropriate) participate in other
related initiatives, in particular to see whether mappings can benefit from
work done elsewhere (e.g. “bibutils”,
http://www.scripps.edu/~cdputnam/software/bibutils/) or to contribute to
enhance the facilities provided by collaborative tools such as Zotero
(www.zotero.org; e.g. by writing a converter to COinS).
New features – evolution of the TEI guidelines
It is expected that the activities planned in this SIG will lead to a
thorough assessment of the biblStruct element as it stands in the current
version of the guidelines, and will probably lead to some feature requests
to the TEI Council. Specific topics already identified relate to the proper
representation of affiliations (org – orgName), typology
of publication objects (biblStruct/@type), accurate representation of
publishing instances (publisher – pubPlace) and
constraints on embedded works (“scoping” relatedItem, good
practices on embedding bibliographic items).
Some other new features to be discussed will be the possibility to use
textClass as a child of monogr, analytic or
series, in order to encode keywords for publications, the
encoding of copy-specific identifiers like shelfmarks or repositories
(following the structure of msIdentifier perhaps as
exIdentifier or copyIdentifier), the use of
biblScope as a direct child of monogr and
series (it’s use in imprint seems somewhat improper)
and the possibility to use extent in biblStruct and the
distinction of several types of extents (eg. pages, sentences and words),
either via an @type-Atribute (which is currently not allowed for
extent oder by using different types of measure in the
extent-Tag.
Potential participants
ChristianeFritze<orgname>Berlin-Brandenburg Academy of Sciences</orgname>BerlinOliverDuntze<orgname>Berlin-Brandenburg Academy of Sciences</orgname>BerlinAndreas Nolda<orgname>Humboldt University</orgname>BerlinLaurent Romary,<orgname>INRIA-Gemo</orgname>BerlinGodfried Croener<orgname>University of Liverpool</orgname>Michel Dalmau<orgname>Indiana University Digital Library Program</orgname>Martin Holmes<orgname>Univ. of Victoria</orgname>Patrice Lopez<orgname>European Patent Office</orgname>,
BerlinMelanie Schlosser<orgname>Ohio State University</orgname>Werner Wegstein<orgname>Univ. of Würzburg</orgname>Grace Wiersma<orgname>MIT Libraries</orgname>,
BostonKevin Hawkins<orgname>Univ. of Michigan</orgname>Hans-Günter Schmidt<orgname>Univ. of Würzburg</orgname>Wendell Piez<orgname>Mulberry Tech.</orgname>Melissa Terras<orgname>University College London</orgname>Julia Flanders<orgname>Brown University</orgname>
