This page describes rules for datasets, conformance to which enables datasets to be integrated in the Loc-I index.
Loc-I focuses on spatial indexing and linking.
Loc-I is concerned with descriptions of geospatial entities, known as features, which are described within datasets that are maintained by distinct authorities and processes. The goal is to allow the spatial relationships with features from other datasets to be easily accessible - particularly spatial containment and overlap. While these relationships might in theory be computed on-the-fly, for various reasons this is not always practical or desirable. The Loc-I index computes and assembles the relationships in advance, through various procedures which are tuned to the different sources. These relationships then support the various Loc-I services, such as Excelerator and IderDown
Each significant relationship between features is realised as an explicit ‘link’. Metadata is associated with each link in order to indicate where it comes from. The set of links is managed and stored separately from either the source or target data.
To support Loc-I, access to a linked-data view of each Loc-I dataset through an RDF representation of the data is required.
Properties required for a spatial dataset to be compatible with Loc-I:
Useful but optional properties:
All other feature properties are application specific, thus not normally of direct interest in the context of Loc-I.
The Loc-I core feature model uses elements from the following standard RDF vocabularies
||http://purl.org/dc/terms/||Dublin Core terms|
The Loc-I core feature model implements the mandatory and option properties as follows:
rdf:typeif the classifier is an
owl:Class, which must be a sub-class of
dcterms:typeif the classifier is something else, such as a
Core model for Loc-I features. Required elements in bold. Application-specific classes and properties in grey
RDF datasets may be serialized in many ways. The Terse RDF Triple Language (Turtle) is a lightweight format that is easily editable and quite readable. JSON-LD is a JSON-based format that might be more usable in many web contexts.
An RDF dataset may be tested for conformance to the Loc-I requirements in many ways. The Shapes Constraint Language (SHACL) provides a standard, executable method. SHACL playground is a convenient tool for basic testing. pySHACL is a
The constraints on
geo:Feature required for Loc-I compatibility are defined in a
loci:Feature in the Shapes graph designed for for Loc-I. This can be used to test data for basic conformance to Loc-I requirements.
The significant benefit of having a common core is that the queries required for standard Loc-I functionality are also common across datasets - i.e. the SPARQL queries will have no dependence on RDF predicates from application-specific vocabularies, only from
(detailed examples TBC)
<http://linked.data.gov.au/dataset/asgs2016/statisticalarealevel2/205031088> rdf:type geo:Feature ; dcterms:identifier "205031088"^^asgs-id:sa2Maincode2016 ; rdfs:label "French Island"^^asgs-id:sa2Name2016 ; rdf:type asgs:StatisticalAreaLevel2 ; geo:hasGeometry <http://gds.loci.cat/geometry/asgs16_sa2/205031088> ; geox:hasAreaM2 [ data:value 170229100.0 ; ] ; geo:sfContains <http://linked.data.gov.au/dataset/asgs2016/statisticalarealevel1/20503108801> ; geo:sfWithin <http://linked.data.gov.au/dataset/asgs2016/stateorterritory/2> ; geo:sfWithin <http://linked.data.gov.au/dataset/asgs2016/statisticalarealevel3/20503> ; loci:isMemberOf <http://linked.data.gov.au/dataset/asgs2016/statisticalarealevel2/> ; . <http://linked.data.gov.au/dataset/asgs2016/meshblock/20663970000> rdf:type geo:Feature ; dcterms:identifier "20663970000"^^asgs-id:mbCode2016 ; rdf:type asgs:MeshBlock ; dcterms:type asgs-cat:primary-production ; geo:hasGeometry <http://gds.loci.cat/geometry/asgs16_mb/20663970000> ; geox:hasAreaM2 [ data:value 58387600.000000007450580596923828125 ; ] ; geo:sfWithin <http://linked.data.gov.au/dataset/asgs2016/stateorterritory/2> ; geo:sfWithin <http://linked.data.gov.au/dataset/asgs2016/statisticalarealevel1/20503108801> ; loci:isMemberOf <http://linked.data.gov.au/dataset/asgs2016/meshblock/> ; . <http://linked.data.gov.au/dataset/gnaf-2016-05/address/GAVIC411436309> rdf:type geo:Feature ; dcterms:identifier "GAVIC411436309"^^gnaf:gnaf-2016-05 ; rdfs:label "Address GAVIC411436309 of Rural type" ; rdf:type gnaf:Address ; dcterms:type <http://gnafld.net/def/gnaf/code/AddressTypes#Rural> ; geo:hasGeometry [ a sf:Point ; gnaf:gnafType <http://gnafld.net/def/gnaf/code/GeocodeTypes#PropertyAccessPointSetback> ; dcterms:type <http://gnafld.net/def/gnaf/code/GeocodeTypes#PropertyAccessPointSetback> ; geo:asWKT "<http://www.opengis.net/def/crs/EPSG/0/4283> POINT(145.35714361 -38.34785008)"^^geo:wktLiteral ; rdfs:label "Property Access Point Setback" ; ] ; loci:isMemberOf <http://linked.data.gov.au/dataset/gnaf-2016-05/address/> ; . <http://linked.data.gov.au/dataset/geofabric/drainagedivision/9400210> rdf:type geo:Feature ; dcterms:identifier "9400210"^^geof:geofabric-id ; rdf:type geof:DrainageDivision ; geo:hasGeometry <http://gds.loci.cat/geometry/geofabric2_1_1_awradrainagedivision/9400210> ; geox:hasAreaM2 [ data:value 134617156547.115 ; <http://www.w3.org/ns/qb4st/crs> <http://www.opengis.net/def/crs/EPSG/0/3577> ; ] ; geox:hasAreaM2 [ data:value 135039327241.81703 ; <http://www.w3.org/ns/qb4st/crs> <http://www.opengis.net/def/crs/EPSG/0/4938> ; ] ; loci:isMemberOf <http://linked.data.gov.au/dataset/geofabric/drainagedivision/> ; .
As mentioned above, links between features are realised and stored explicitly. See this explanation of the schema for link statements.
An Application Schema is a data model designed to support the functional requirements of a particular application. This will involve more specific classes and properties related to the domain of discourse. Design of a application-schema is the responsibility of the data provider, in consultation with their user community.
In order to ensure that an Application Schema complies with the conformance rules above, it is recommended to use Loc-I core as the basis, and add additional classes and properties as required to support the application. Some of the classes in an application schema will be Features, and these should be axiomatized as sub-classes of
geo:Feature. Other classes and properties will extend the application schema outside the scope of Loc-I, and there are no specific constraints arising from Loc-I in these cases.
See Loc-I application schemas for a basic outline of the ontologies for the initial Loc-I conformant datasets.