I apologize for not chiming in sooner ...
The metadata mapping work, I believe, falls nicely under the Libraries
SIG purview, and we are engaging in
some of this work as it relates to the overhaul/update we are doing to
the "TEI Text Encoding in Libraries
Guidelines for Best Encoding Practices" (http://www.diglib.org/standards/tei.htm
). Currently, we have a team of two
folks, Kevin Hawkins from Michigan and Melanie Schlosser from Ohio
State University, working on documenting best
practices for the TEI Header including referencing other metadata
formats such as MARC. The SIG has also discussed
amassing documentation, tools, etc. related to mapping, crosswalks and
so on from the Header to MODS, DC and other
Also, we are not assuming that the TEI Header should be treated only
as the derivative bibliographic format
for Digital Library-related projects. Often, we are creating "hand-
tailored" metadata, perhaps not for printed texts with
MARC records, though this is not entirely true such as the case with
serials, but for manuscripts, legal and
administrative documents, and other text documents often encoded in
the library context. We are trying to
make sure we cover as many scenarios as possible when updating the TEI
Header section of the "Guidelines
for Best Encoding Practices" document.
Recently, Matt Gibson from University of Virginia and I co-chaired a
joint DLF TEI Task force and Libraries SIG meeting that was
hosted by the Digital Library Federation during their Spring Forum in
Minneapolis. I will soon be posting a summary
of the meeting and meeting minutes for public review and feedback.
The Libraries SIG's primary work, as I mentioned earlier, is to
update the TEI Text Encoding Guidelines and migrate the Guidelines
under the custodianship of the SIG.
We will be forming working groups around this initiative for broader
contributions because we need help too. Perhaps we can
expand the scope of the "TEI Header" group to also tackle greater
metadata issues. However, if a specialized SIG forms, which
is also an option as already proposed, then the Libraries SIG and this
new SIG should make a point to stay abreast of each other's
work to minimize possible redundancies.
While here, I also wanted to comment on the original biblStruct v.
biblFull post as it relates to bibliographic metadata for serials.
At Indiana University, we just launched a TEI-encoded 102-year run of
a scholarly journal, the Indiana Magazine of History,
and we encountered a few challenges in representing article-level (and
even more granular-level metadata for book reviews)
using the TEI. Our issues are probably complementary to Oliver's
struggle especially if we want to capture robust bibliographic
metadata using the TEI. For us, it was important to have an
authoritative form (TEI) of both the document markup and the metadata.
Because we used P4, we solved this by using the TEI Independent Header
to capture granular metadata although, based on an informal
survey we circulated, the Independent Header is more commonly used to
capture series-level metadata.
Capturing metadata for serials made it as an agenda item at the last
TEI Members' Meeting, but
we didn't have enough time to explore all the issues. I suspect this
topic will eventually surface for discussion, as it has here, via the
list or in future meetings.
So stay tuned for the meeting summary update and the working groups
that will emerge to meet some immediate goals identified
by the Libraries SIG including the Header best practices
documentation. Perhaps we can cultivate synergies or expand working
to take on a greater scope.
--Michelle, Chair of the Libraries SIG
| Michelle Dalmau, Digital Projects & Usability Librarian
| Indiana University Digital Library Program
| Herman B Wells Library
| 1320 East 10th Street, W501
| Bloomington, Indiana 47405
| (812) 855-1261, [log in to unmask]
On May 19, 2008, at 6:57 PM, Lou Burnard wrote:
> Rather than a work group, I think a SIG would be exactly the right
> mechanism for doing this. Bravo for having suggested it -- and good
> Now, who else wants to stand up and be counted? (I'm looking at some
> German people with broomsticks ....)
> I agree with you that, although the existing Libraries SIG obviously
> has a great deal of relevant expertise to contribute, and I'm sure
> some of its members would wish to contribute to this effort, its
> primary area of activity is somewhat different.
> The mechanism for setting up SIGs is documented on the website at http://www.tei-c.org/Activities/SIG/rules.xml
> though I'm not sure how up to date this page is....
> Grace Wiersma wrote:
>> Hi Lou (and others interested),
>> Would it be possible to establish a working group on this topic, to
>> take the issue out of the realm of belief and into a more empirical
>> realm? I don't mean to be jocular--this is an issue that's come up
>> before and about which I've corresponded with Martin Holmes in the
>> past. We discussed pulling together available mappings between MARC
>> and other metadata standards, then creating a "microformat" (and
>> related GUI tool) for getting the right stuff into the TEI Header
>> without each project having to re-invent the wheel. The problem so
>> far (at least for me) has been to find time to work on it. With a
>> more public working group/discussion devoted to it, perhaps there'd
>> be enough structure and encouragement for those interested to get
>> work done and report on it. I see from this discussion the
>> beginnings of such a group. Or, perhaps such a group already exists
>> and we haven't heard from them yet?
>> My sense is that the TEI in Libraries SIG already has enough on its
>> plate, and that after all its focus is upon issues of metadata
>> automation, since the mainstream work of libraries nowadays is not
>> hand-tailored metadata for individual texts, but getting the
>> available standards to converge and work to scale, within the
>> context of the large automated projects for which libraries are
>> typically responsible.
>> Grace Wiersma
>> Cataloging & Metadata Services
>> MIT Libraries
>> [log in to unmask]
>> (617) 253-0643
>> -----Original Message-----
>> From: TEI (Text Encoding Initiative) public discussion list [mailto:[log in to unmask]
>> ] On Behalf Of Lou Burnard
>> Sent: Monday, May 19, 2008 10:52 AM
>> To: [log in to unmask]
>> Subject: Re: biblFull or biblStruct
>> Well I haven't checked, but my belief is that there are examples of
>> how to do most of these (using biblStruct) in the current
>> guidelines. Discussion on this thread seems to have been led astray
>> by the assumption that biblFull could do everything biblStruct
>> does. Which it doesnt.
>> That's not to say that it's impossible to come up with complex
>> situations which the Guidelines don't cater for, and which would
>> also probably defeat most other existing cataloguing systems, but I
>> don't think that warrants the suggestion that the Guidelines are
>> incapable of supporting library standard bibliographic
>> descriptions, for which there is ample existence proof to the