This is the pleiades.rdf package. It provides RDF views of individual Pleiades places (such as http://pleiades.stoa.org/places/1043/turtle) and the comprehensive RDF dataset dumps at http://atlantides.org/downloads/pleiades/rdf/.

The RDF views of Places (such as http://pleiades.stoa.org/places/1043/turtle for Turtle or http://pleiades.stoa.org/places/1043/rdf for RDF/XML) contain triples with the following classes of subjects:

• Ancient World Places (real past world entities)
• Authors – foaf:Person
• Pleiades Places, Names, Locations – pleiades:Place, pleiades:Name, pleiades:Location
• Pleiades vocabulary items – skos:Concept

Generally, every thing in the graph has at least a label, but detailed data is provided only for the Place at the center of the graph and its Locations and Names. The vocabularies used to express the detailed data are enumerated below.

The pleiades-latest.tar.gz archive contains 9 files of RDF for Places (places-[1-9].ttl), one file of RDF about erroneous Places (errata.ttl), a file describing Pleiades authors (authors.ttl), and two vocabularies (place-types.ttl and time-periods.ttl).

These archives are created weekly. The difference between these files and the single Place graphs is that authors and vocabularies are represented separately and not duplicated within the batches of Places.

Pleiades uses terms from several different vocabularies and ontologies.

http://purl.org/spar/cito/

The terms cito:cites, cito:citesAsRelated, cito:citesForEvidence, and cito:citesForInformation are used to identify works cited and their context.

http://purl.org/dc/terms/

Every content item in Pleiades has a dc:title, dc:description, dc:creator, and one or more dc:contributor.

http://xmlns.com/foaf/0.1/

All content creators and contributors are foaf:Person.

http://geovocab.org/spatial#

Geographic connections are expressed using spatial:C.

http://data.ordnancesurvey.co.uk/ontology/geometry/

http://data.ordnancesurvey.co.uk/ontology/spatialrelations/

The extents of spatial objects are expressed using osgeo:AbstractGeometry, osgeo:AsGeoJSON, and osgeo:AsWKT. Spatial overlap is expressed using osspatial:partiallyOverlaps.

http://www.w3.org/2002/07/owl#

The owl:sameAs property is used to identify the resolved duplicates within Pleiades.

http://pleiades.stoa.org/places/vocab#

The Pleiades RDF vocabulary consists of 3 classes: pleiades:Place, pleiades:Location, and pleiades:Name.

Instances of pleiades:Place identify their component Names and Locations with pleiades:hasName and pleiades:hasLocation.

All Names and Locations identify time periods during which they were in use with pleiades:during. The minimum attested year of the oldest and the maximum attested year of the most recent of these time periods are also provided using pleiades:start_date and pleiades:end_date.

Locations identify associated feature types with pleiades:hasFeatureType.

The attested form of a name in its original writing system is bound to the pleiades:nameAttested property, while its romanizations are bound to pleiades:nameRomanized.

http://www.w3.org/2004/02/skos/core#

SKOS terms are used to describe the feature type and time period vocabularies of Pleiades.

http://www.w3.org/2000/01/rdf-schema#

We try to give every thing in our graph a rdfs:label. Things of the ancient world – as opposed to their counterparts in the Pleiades site – have a rdfs:comment instead of dc:description. Pleiades also uses rdfs:seeAlso for links to Wikipedia, etc.

http://www.w3.org/2003/01/geo/wgs84_pos#

This is probably the most well known vocabulary for geographic location. The terms geo:lat and geo:long are used to help spatially naive systems get a grip on Pleiades data.

http://www.w3.org/TR/prov-o/#

The sources of all Pleiades content are identified using prov:wasDerivedFrom.

See the file pleiades/rdf/cap-authors.csv.