srda provides human readable and meaningful lexical equivalents for all RDA elements, defined using common owl:sameAs relations.
RDA, Resource Description and Access, is a modern metadata standard designed for use in the bibliographic domain. Unfortunaltely, the commission that designed RDA chose not the use meaningful names for the RDA elements but abstract codes instead.
Do you think the following RDF is easy to understand? :
@prefix rdaw: <http://rdaregistry.info/Elements/w/> .
<id/som/work/5882fb21c4a696cc877f59e910c81fca>
rdf:type <http://rdaregistry.info/Elements/c/C10001> ;
rdaw:P10223 "Chamber music" ;
rdaw:P10004 <id/som/d78168> ;
rdaw:P10053 <id/som/7160833597948a09eb41d51c99eb8de2> ;
rdaw:P10061 <id/som/6e6568e39715aad83d5dac96464ec33a> ;
rdaw:P10069 "Delta Ensemble" ;
rdaw:P10219 "1983" ;
rdaw:P10220 <id/som/um2532> , <id/som/um914> ;
rdaw:P10287 "Amsterdams Fonds voor de Kunst" ;
rdaw:P10353 <https://iso639-3.sil.org/code/eng> .
Or do you consider this equivalent RDF to be more human readable? :
@prefix srda: <https://data.digitopia.nl/srda#> .
<id/som/work/5882fb21c4a696cc877f59e910c81fca>
rdf:type srda:Work ;
srda:preferredTitleOfWork "Chamber music" ;
srda:authorAgent <id/som/6e6568e39715aad83d5dac96464ec33a> ;
srda:categoryOfWork <http://rdaregistry.info/termList/RDATerms/1118> ;
srda:commissioningAgent "Amsterdams Fonds voor de Kunst" ;
srda:composerAgentOfWork <id/som/7160833597948a09eb41d51c99eb8de2> ;
srda:dateOfWork "1983" ;
srda:dedicateeAgentOfWork "Delta Ensemble" ;
srda:languageOfRepresentativeExpression
<https://iso639-3.sil.org/code/eng> ;
srda:mediumOfPerformanceOfMusicalContentOfRepresentativeExpression
<id/som/um914> , <id/som/um2532> .
For those who prefer the latter, srda has been created.
Working with the abstract element names as provided by RDA, is inefficient and error prone.
To overcome this, srda provides human readable and meaningful lexical equivalents for all RDA elements, defined using owl:sameAs relations.
Additionally, srda provides all the rdf:label properties.
The element names of srda have been automatically derived from the lexical aliasses in RDA (excluding the language suffix). srda is based on the English language, the lingua franca of the world of information systems.
srda lives at https://data.digitopia.nl/srda# (which is also the namespace for srda).
This repository (https://github.com/renevoorburg/srda) provides the scripts that are used to create https://data.digitopia.nl/srda#. Actually, only srda_virtuoso.sparql is used, the script srda_construct.sparql provides an equivalent using a CONSTRUCT query for reference and conveniance.
These queries expect the named graph <http://rdaregistry.info/Elements/v5.0.19/> to be available and loaded with all regular official RDA element definitions. These can be downloaded from https://github.com/RDARegistry/RDA-Vocabularies/releases. Note that the object, datatype, unconstrained and 'rof' definitions should not be loaded. They are not a part of srda (adding them would have resulted in conflicting and duplicate element names).
Indeed, RDA does provide so called lexical aliases. Those lexical aliases are even used to create srda, so why not use them in your RDF?
The problem with the lexical aliases as provided by the official RDA definitions is multifold:
- The property used to define lexical aliases,
<http://metadataregistry.org/uri/profile/regap/lexicalAlias>is not an internet standard, - The property is not defined properly, looking it up results in a
http-404error. - Its semantics are unclear.
Using owl:sameAs, as srda does, fixes these shortcomings.