Use case #10 - Using GRDDL to indicate serializations of
RDF
AlphaCorp publishes and exchanges XML data documents for use by a
variety of internal and external applications. These XML
documents are expressed in a large variety of legacy formats that
cannot be changed. AlphaCorp is frustrated because:
- Even though many of these document formats make use of the same
or similar concepts (such as a mailing address), they are expressed in
different schemas, and thus it is difficult for software to
automatically detect similarities and relationships across document
types.
- Different document types require different kinds of XML
processing to be interpreted correctly, and documentation on the
meaning and correct processing of each document type is often
imprecise or non-existent. This causes persistant problems as
developers implementing new applications (both internal and external)
that consume these documents interpret them differently.
- Several newer RDF applications (both internal and external) would
like to be able to make use of the information in these documents, but
they currently cannot, because the documents are in plain XML, not RDF.
Amy, a software architect at AlphaCorp, reads that GRDDL is used "to
complement the concrete RDF/XML
syntax with a mechanism to relate other XML syntaxes . . . to the RDF
abstract syntax" (GRDDL
charter). She decides to use GRDDL to indicate how each XML
document can be transformed into RDF that captures the entire meaning
of the document, thus allowing the same XML document to be used
interoperably by both XML and RDF
applications. Since RDF is syntax independent, each XML document would be a serialization
of RDF, with a GRDDL transformation indicating the appropriate
deserialization.
Amy persuades the owners of the various document formats (the namespace
owners) that this is a good idea, and then works
with them to write the appropriate GRDDL transformations. After a
few
iterations, the namespace owners decide that they are happy with the
GRDDL transformations, and decide (as a matter of policy) to consider
the resulting RDF to be authoritative regarding the
information content of the XML document. The XML documents then
start
making use of the GRDDL transformations, and AlphaCorp begins
a new era in which XML documents be used with identical meaning by both
XML and RDF applications.
Key considerations in this use case
- As RDF deserialization algorithms, Amy's GRDDL transformations
must be unambiguous: for a given XML
document, it must be crystal clear, exactly what RDF graph that XML
document denotes.
- There is a large variety of document formats -- some using the
XML Infoset and some not -- and different document types require
different XML parsing and pre-processing policies in order to be
interpreted correctly.
- Although Amy understands each document format, and the format
owner
endorses Amy's GRDDL transformations, Amy cannot modify the document
formats.
