Differences

This shows you the differences between two versions of the page.


details:rdfxopts [2023/06/07 20:39] (current) – created - external edit 127.0.0.1
Line 1: Line 1:
 +====== RDF Export options ======
 +
 +RIMMF3 supports two types of RDF Export:
 +
 +  * RDF 'Simple'
 +  * RDF 'Rimmf'
 +
 +The second case, RDF 'Rimmf', supports an export to RDF that is intended to be as close to 100% RIMMF-compatible as possible. This means data exported using this option may be imported back into RIMMF, with no loss of application-specific information. This type of export is always applied to single record exports, and RDA 'set' exports.
 +
 +The first case, RDF 'Simple', is designed for using data generated by RIMMF outside of RIMMF. This type of export includes a number of options that support removing application-specific, and/or semantically unncessary, data from the resulting RDF. This type of export (and the options listed below) is applicable when using the Export tool in the Entity Index.
 +
 +The options for the RDF 'Simple' export are found on the 'Data Options' page (under 'Options' on the main menu):
 +
 +{{:details:rdfxo.png|}}
 +
 +By default, the first two options are selected. A description of each follows.
 +
 +__Reifications__
 +
 +Reifications are used to carry data that the application uses to record user choices about the data in an entity record (ie application-specific data). For example: the order of the elements in an authorized access point, or the fact that an authorized access point contains plain text instead of a link to another resource.
 +
 +Here is an example:
 +
 +  <http://rimmfdata.com/r/{05294B17-0DFE-413C-A181-0C537795B61C} <http://www.w3.org/1999/02/22-rdf-syntax-ns#type <http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement .
 +  <http://rimmfdata.com/r/{05294B17-0DFE-413C-A181-0C537795B61C} <http://www.w3.org/1999/02/22-rdf-syntax-ns#subject <http://rimmfdata.com/r/clinky00000881 .
 +  <http://rimmfdata.com/r/{05294B17-0DFE-413C-A181-0C537795B61C} <http://www.w3.org/1999/02/22-rdf-syntax-ns#predicate <http://rdaregistry.info/Elements/a/c .
 +  <http://rimmfdata.com/r/{05294B17-0DFE-413C-A181-0C537795B61C} <http://www.w3.org/1999/02/22-rdf-syntax-ns#object "1817 July 18" .
 +  <http://rimmfdata.com/r/{05294B17-0DFE-413C-A181-0C537795B61C} <http://rimmf.com/vocab/hasAapComponentFlag "3" .
 +  
 +All of this is needed to 'say something about' a single triple.
 +
 +__Triples that confuse the resource with the description__
 +
 +As an application that purports to be useful to catalogers, RIMMF generates and maintains metadata that catalogers will find useful, like the date an entity record is created, the last modification of a record, etc. However, the way that RIMMF currently exports this data as RDF is semantically confusing:
 +
 +  <http://rimmfdata.com/r/qpq00000087> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://rdaregistry.info/Elements/c/C10007> .
 +  <http://rimmfdata.com/r/qpq00000087> <http://purl.org/dc/terms/created> "2014-04-29T16:50:03"^^<http://www.w3.org/2001/XMLSchema#dateTime> .
 +
 +The subject of the first triple is the resource described by the RIMMF record, but the subject of the second triple is intended to be the RIMMF record itself. However, as it stands, these triples give the creation date of the Manifestation, not the RIMMF record.
 +
 +We are working on a way to provide semantically correct provenance data. In the interim, if you are not planning on importing the RDF back into RIMMF, you should probably exclude these type of triples.
 +
 +__Triples for RIMMF identifiers__
 +
 +By default, these triples are included in the 'RDF (Simple)' export--
 +
 +  <http://rimmfdata.com/r/beast00000110> <http://rimmf.com/vocab/rimmfIdentifierFamily> <http://rimmfdata.com/r/beast00000110> .
 +
 +--but if you do not need or want them you can easily turn them off here. Note that and all elements from http://rimmf.com/vocab/, like http://rimmf.com/vocab/rimmfIdentifierFamily, support dereferencing.
 +
 +
 +__Triples that contain entity record labels__
 +
 +We were taught that it was good practice to include human-readable labels for things in our linked data; and by default, these triples are included in the 'RDF (Simple)' export--
 +
 +  <http://rimmfdata.com/r/beast00000596> <http://www.w3.org/1999/02/22-rdf-syntax-ns#label> "Austen, Jane, 1775-1817" .
 +
 +--but since they are the same as the authorized access points generated by the program
 +
 +  <http://rimmfdata.com/r/beast00000596> <http://rimmf.com/vocab/authorizedAccessPointPerson> "Austen, Jane, 1775-1817" .
 +  
 +--you may want to turn them off. We may develop a map from the RIMMF element set to rdfs:label so that these triples can be inferred.
 +
 +
 +__Triples that can be inferred from RDA constraints__
 +
 +RIMMF uses the RDA Registry constrained elements. Thus, consumers of RIMMF data may infer the class of the entity record from the properties exported. For example, these triples describe an expression-- 
 +
 +  <http://rimmfdata.com/r/beast00000597> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://rdaregistry.info/Elements/c/C10006> .
 +  <http://rimmfdata.com/r/beast00000597> <http://rimmf.com/vocab/authorizedAccessPointExpression> "Austen, Jane, 1775-1817. Northanger Abbey. French" .
 +  <http://rimmfdata.com/r/beast00000597> <http://rdaregistry.info/Elements/e/P20002> "LCCN n 2013029805" .
 +  <http://rimmfdata.com/r/beast00000597> <http://rdaregistry.info/Elements/e/P20006> "French" .
 +
 +But the class of expression can be inferred from the use of the expression elements  
 +
 +  <http://rdaregistry.info/Elements/e/P20002>
 +  <http://rdaregistry.info/Elements/e/P20006>  
 +
 +because every property in that expression element set has a domain of Expression (rdac:C10006)
 +
 +"Applications which need to follow schema and ontologies, such as RDA/FRBR and BIBFRAME, should 
 +be aware of domain classes and types, but other applications can safely ignore them. This is so-called 'ontological commitment' - something which RDA must adhere to if the power of the WEMI relationships is to be leveraged.
 +So it is better to allow the application to make and use the entailment, and not explicitly include it in the RDA 'record'."--Gordon Dunsire
 +
 +
  
Back to top
CC Attribution-Share Alike 4.0 International
Driven by DokuWiki