Publishing Linked Data - Linked Data: A Geographic Perspective - page 124

RDF. It includes some of the same information as the Semantic Sitemap (such as
the data dump and SPARQL endpoint) but also describes the vocabulary used in
the dataset and its outgoing and incoming links, as well as specifying any subsets
of the dataset.
There are two main ways of encoding metadata in RDF: reification and Named
Graphs. As we saw in Chapter 5, the technique of reification specifying an
rdf:Statement with a subject, predicate, and object allows other information,
namely, metadata, to be added about the rdf:Statement . There are several draw-
backs to this approach, however. First, it makes querying difficult as the triple graph
model has been broken. Second, it buries metadata within the data itself, potentially
making the volume of data balloon and forcing a user to deal with both data and
metadata at the same time. VoID takes an alternative approach, to assign a URI
to the graph itself, and then metadata about the graph can be also assigned to this
URI. The graph is then known as a Named Graph as it has been given a URI. This
approach maintains the RDF triple structure throughout both the data and metadata,
and the metadata can conveniently be kept separate from the main data. The down-
side is that a system of unique identifiers must be maintained for every named graph
in the dataset, which becomes more of a management issue as the granularity of the
metadata becomes finer. For example, while it is straightforward enough to create
one URI for the whole dataset and assign metadata to the whole dataset, if we wish
to add different metadata to each triple (which might be particularly useful if the
triples have different provenance), we will need to provide a separate URI to each
“Named Triple.”
The VoID vocabulary defines the class void:Dataset , an instance of which is
an RDF resource that represents the whole dataset, so that statements can be made
about the whole dataset's triples. VoID also describes linksets, collections of RDF
links between two datasets. A link is an RDF triple whose subject is in one dataset
and object in another. The void:Linkset class is a subclass of void:Dataset.
The following shows an example of VoID metadata for Merea Maps' administrative
geography RDF dataset:
1. <?xml version = “1.0”?>
2. <rdf:RDF xmlns:foaf = http://xmlns.com/foaf/0.1/
3. xmlns:void = http://rdfs.org/ns/void/
4. xmlns:rdf = http://www.w3.org/1999/02/22-rdf-syntax-ns#
5. xmlns:dcterms = “ http://purl.org/dc/terms/” >
6. <void:Dataset rdf:about = “ http://data.mereamaps.gov.me” >
7. <foaf:homepage rdf:resource = “ http://page.mereamaps.gov.me”/ >
8. <dcterms:title>Administrative Geography</dcterms:title>
9. <dcterms:description>An example Merean dataset in
10. RDF.</dcterms:description>
11. <dcterms:publisher rdf:resource = “ http://mereamaps.gov.me”/ >
12. <dcterms:license
13. rdf:resource = “ http://www.opendatacommons.org/licenses/
pddl/1.0/”/ >
14. <void:sparqlEndpoint rdf:resource = “ http://mereamaps.gov.me/
sparql”/ >