[Distribution] Update and clarify usage of mobilityDataStandard property and MobilityDataStandard

[marioscrock]

Result of the discussion of the dedicated meeting held on Jan 8, 2026 (cf. #148).

Issues on the topic (now closed and summarised here):

• #73 Clarify that mobilityDataStandard is subclass of conformsTo (used in DCAT-AP) as it restricts the range w.r.t. DCAT-AP
• #93 Add property linked schema to Distribution
• #65 Expected usage of property mobilityDataStandard and new property conformsTo

Note 1: the mobility data standard is an information attached to Distribution (and not Dataset) following the modelling in DCAT where multiple Distributions, serialised using different data formats/standards (e.g., GTFS and NeTEx) can be associated with the same Dataset (e.g., timetables of an operator). In DCAT-AP, it is allowed to define conformsTo at the Dataset level, however, this is not considered at the moment in mobilityDCAT-AP. If someone wants to also add conformsTo at the Dataset level, please create a new issue and provide a proper justification.

Note 2: in DCAT-AP the property dct:conformsTo is referred with the label linked schema but these are not used as separated properties.

Clarify mobilityDataStandard is a sub-property of dct:conformsTo

In mobilityDCAT-AP, the property mobilityDataStandard should be defined as a sub-property of dct:conformsTo, and MobilityDataStandard as a subclass of dct:Standard (see link for more information from SEMIC guidelines). Indeed, mobilityDataStandard is made mandatory, and we restricted the range to MobilityDataStandard. Moreover, MobilityDataStandard is defined as a new class with properties dct:conformsTo, owl:versionInfo and mobilitydcatap:schema.
In any case, dct:conformsTo remains admissible for use with other values (given that this is defined in DCAT-AP), but it is not currently explicitly mentioned as a property of Distribution in mobilityDCAT-AP.
From a "logical" perspective, a subproperty should always be interpretable as the corresponding super-property. "Practically", it may happen that the person harvesting the metadata is not applying the "inference" necessary to interpret and retrieve mobilityDataStandard values together with the dct:conformsTo ones. However, this is not something to be solved at the specification level.

Update and clarify usage

Example of the proposal agreed upon during the meeting. It would be the same as saying that this Distribution conforms to NeTEx but also conforms to this profile of NeTEx (but also conforms to this specific schema, etc.). This will facilitate searching for Distributions conforming to the same standard (using values from our controlled vocabulary) and without breaking compatibility.

<https://www.example.com/dataset/234r2efhbn/resource/akjsdefh298>
  a dcat:Distribution ;
  [...]
  mobilitydcatap:mobilityDataStandard <https://w3id.org/mobilitydcat-ap/mobility-data-standard/netex> , 
    <https://www.example.com/standard/netex-italian-profile-4.2.0> .

<https://www.example.com/standard/netex-italian-profile-4.2.0>
  a mobilitydcatap:MobilityDataStandard ;
  dct:title "NeTEx Italian Profile version 4.2.0" ;
  dct:conformsTo <https://w3id.org/mobilitydcat-ap/mobility-data-standard/netex>,
   <https://raw.githubusercontent.com/5Tsrl/netex-italian-profile/refs/heads/main/xsd/NeTEx_publication_Lev5.xsd> ;
  owl:versionInfo "4.2.0" .

Additional values for mobilityDataStandard could be added, for example, to specify that the schema used by this Distribution also conforms to a specific subset of the NeTEx Italian Profile v4.2.0 (e.g., https://www.example.com/standard/netex-italian-profile-4.2.0-fare-frame).

Changes to be applied if accepted:

• mobilityDataStandard should be defined as a sub-property of dct:conformsTo in the Respec page, as it is not mentioned currently
• Add linked schema serialised as dct:conformsTo to Distribution as in DCAT-AP
• The RDF serialisation of mobilityDataStandard should be fixed as now it is rdfs:subPropertyOf dct:format
• Increase cardinality of mobilityDataStandard to 1..*
• MobilityDataStandard has a property standard name currently mapped to conformsTo that is confusing. If we want to keep a property for the name, serialise as dct:title. Rename conformsTo to conforms to or linked schema.
• The property mobilitydcatap:schema is superfluous and can be removed.
• Recommend the provision of at least one value from the mobility-data-standard controlled vocabulary (noting that also other is available) even if a new individual of MobilityDataStandard is provided with additional information
• Provide additional documentation (on the Wiki) on the different options available to provide additional details on the format of the Distribution and how to facilitate harvesting in open data portal. Add also a note for NAPs to reuse where possible IRIs of standards defined (e.g., avoid defining different IRIs for each Distribution referencing a MobilityDataStandard corresponding to the same version of the Italian NeTEx profile)