Usage guidelines and examples

(DRAFT) The main objective of this page is to provide a provenance publisher the basic idea of what is needed to apply our domain model.

(I have started with RDF because is the most common language in the Semantic Web. Other sections like this one can be created).

• Grouping statements: before incorporing the model to a new domain it is mandatory to analyze which are the different descriptions levels: how can we group the descriptions in different layers in order to be able to describe them properly.
• The proposed domain model may not be necessary if no statements about description sets exist. However, it is recommendable to group the descriptions at least in one level (in case that other entities aggregate the published information in the future).
• 'Other requirements'

(Based on the Use Cases)

El Viajero encoding example (pubby example of the use cases):

'short explanation as summary from the paper': At the Universidad Politécnica de Madrid currently a project is carried out that aims to create a repository of metadata about tourist resources (i.e. guides, images, and videos). Each resource is assigned a different URI, which is used to describe it with its provenance information (creator, date of creation, references used, etc.) as well as additional descriptive metadata about the resource (size, title, subtitle, etc.)

'separating the data in different levels': The metadata was created by a spanish newspaper , but it was made public in RDF by the UPM at a certain date under a certain license, so there is a lot of interest in reflecting that in the RDF created too. The RDF is exposed as Linked Data in a repository accessed via Pubby , a linked data frontend for SPARQL endpoints which allows exploring and navigating through the links of the endpoint. Pubby allows to define an additional level of metadata, since it provides information about the RDF shown to the final user (e.g. the query used to retrieve the RDF from the server, the date of retrieval, the web service used to perform the query, etc.), and it describes it using the Provenance Vocabulary . Therefore we can organize the metadata in three different levels or groups: the first one groups the descriptions about the resource, the second one gathers the descriptions about the previous statements and the last one refers to the RDF serialization of the first two groups, which is what is shown to the users.

'encoding the example in RDF:'

@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>.
@prefix xsd: <http://www.w3.org/2001/XMLSchema#>.
@prefix dc: <http://purl.org/dc/elements/1.1/>.
@prefix dcterms: <http://purl.org/dc/terms/>.
@prefix dcprov: <http://namespaceNotYetKnown/.
@prefix ex: <http://example.org/data/>.
# DescriptionSet1: descriptive statements about a resource.
<http://example.org/data/guideIdentifier/desc1> 
{ 
	ex:guideIdentifier dc:date "2011-05-27"^^xsd:date.
	ex:guideIdentifier dc:creator ex:Paco_Nadal.
	ex:guideIdentifier dc:rights "El País" .
	<http://example.org/data/guideIdentifier/desc1> rdf:type  dcprov:DescriptionSet.  
}
# AnnotationSet1: creator, date and publisher of DescriptionSet1
<http://example.org/data/AnnotationSet/annSet1> 
{ 
	<http://example.org/data/guideIdentifier/desc1>	dc:date "2011-05-28"^^xsd:date.
	<http://example.org/data/guideIdentifier/desc1>	dc:creator ex:Prisa_Digital.
	<http://example.org/data/guideIdentifier/desc1>	dc:publisher ex:UPM.
	<http://example.org/data/AnnotationSet/annSet1>	rdf:type  dcprov:AnnotationSet .
}
<http://example.org/data/AnnotationSet/annSet2> 
{ 
	<http://example.org/data/AnnotationSet/annSet1>	prv:createdBy ex:DataCreation1.
	<http://example.org/data/AnnotationSet/annSet2>	rdf:type  dcprov:AnnotationSet .
}