Schottlaender,
B.E.C. (2003). Why metadata? Why me? Why now?. Cataloging & Classification
Quarterly 36(3/4):19-29.
This article reviewed different types of metadata,
and the various metadata standards of various divergent communities of data and
data compilers that contribute to a confusing multiplicity of metadata
standards. This complexity is demonstrated by the three different types of
metadata standards that can often be interrelated on a single document, for nearly
unlimited combinations of forms. In addition to the ‘normal’ schema of metadata
– that is, how the data is stored – there is an additional schema as to how
that metadata is encoded and displayed. Beyond those two there are other
standards for the architectural schema of metadata, which seems to deal more
with collocation and organization. He emphasized the mutability of e-documents,
their natural flexibility requires more stringent metadata standards to tie
them down so they aren’t lost to the ether.
He also identified two distinct types of organization
schemes for metadata: syntactical and semantic. Most organizational schemes in
the metadata communities are primarily focused on syntax, which is how the
information is laid out, which information is included, and how those things
are tagged and coded. Semantic considerations involve how the actual words and
data are spelled out in terms of shared vocabulary, spelling, and punctuation.
Most schemes rely on librarians to create semantic standards, or simply don’t
have any, utilizing AACR standards and the Library of Congress subject
classification to fill in gaps.
However, many in the metadata community are
starting to come around to the idea that semantics really are important in
terms of searchability, and acknowledging the work and expertise of libraries. One
prominent member of the metadata community stated that the Dublin Core schema
needed to align more with the logic and underlying structure of FRBR, and
needed to look more like library cataloging. My one note of concern, however,
is that sometimes fresh eyes see a problem in a clearer way than those of
experience, and librarians should be careful not to stifle innovative
philosophy under the weight of decades of experience.
I was particularly interested in the
author’s description of architectural schema. Of the many articles we’ve read
for class, I’ve seen many competing types of metadata standards, all of which
are mutually exclusive in terms of collocation. The complexities of this system
demonstrate to me how difficult it can be to find data and resources, and how
this affects the dispersal of information. I can’t search every database in the
world for the correct information when their metadata schemes don’t work
together. Interoperability seems like it should be a priority for both metadata
and library communities. Throughout the readings, I’ve been thinking, ‘someone
needs to come up with a system to read and interpret all the different types of
information storage, so that it can all be made jointly searchable.’
Architectural schema seem to be the
solution to this. The author mentioned the “Storage Resources Broker which is
predicated on the Warwick Framework,” a type of architectural schema. “It is a software
suite that allows one to pull a variety of digital objects into a container architecture
that can handle basically any kind of metadata” (24). That sounds particularly exciting
for the possibilities of multiple system collocation and information retrieval.
No comments:
Post a Comment