Home » railML newsgroups » railml.common » [railML3] Dublin Core metadata update in railML 3.3
Re: [railML3] Dublin Core metadata update in railML 3.3 [message #3661 is a reply to message #3652] Mon, 23 June 2025 15:04 Go to previous messageGo to previous message
Rémi Collet is currently offline  Rémi Collet
Messages: 13
Registered: November 2024
Junior Member
Hello Thomas,

Thank you very much for this update on the use of DCT !

1. I would suggest that super-properties like dct:coverage or dct:date that are accepted but not recommended should not but put in the "stable" part of the document, so it is clear that we recommend using more specific sub-properties. Or perhaps make it clearer that their use is not recommended, by adding a column ?

2. In a broader context, we should also decide what vocabulary set to use for specific areas. The Skos vocabulary (https://www.w3.org/TR/swbp-skos-core-spec/) is a W3C standard, is broadly used, and has some properties that somewhat double what is available in Dublin Core. For example, the dct:alternative property is used to provide an alternative title for a resource. Skos is more precise in that regard and provides more specific properties like skos:altLabel, skos:hiddenLabel, and skos:prefLabel. Do we need to be as precise as Skos, or is Dublin Core fine ?

3. Finally, concerning your main question, if correctly understood it, I am very much in favour of using the more updated version of DCTerms where possible. It is backward compatible, so no worries there.

Best regards,


Ontologist @Infrabel (Belgian Railway Infrastructure Manager)
remicollet(at)infrabelbe
 
Read Message
Read Message
Read Message
Previous Topic: [railML3] Adding globally unique identifiers
Next Topic: [Ontology v0.6] Missing properties for signals
Goto Forum:
  


Current Time: Sun Feb 15 02:40:45 CET 2026