Re: issue-202: asserting the dataset's metadata conforms to certain standards

Hi Riccardo,

Sorry for the late response. I hope someone would chime in, to approve or object...

The way to represent standards themselves, the example for :geoDCAT-AP seems fine (maybe others will have some more input on this).

What puzzle me is the property dqv:hasMetadataConformingTo
It does do the job, but it's quite an indirect statement to attach to a dataset. I mean, what is conformant with GeoDCAT-AP is the metadata, not the dataset itself.
It could be that in another databse, the dataset description wouldn't be conformant.

Of course if the requirement is to indicate something like "somewhere the dataset has a description that complies with GeoDCAT-AP", then dqv:hasMetadataConformingTo is ok.
But I'm feeling this could be only a near-match to a data consumer need. I mean, I think they should get this information only when looking at a specific piece of metadata that represents the dataset, not "in general".
Actually the proposal could be harmful if a data consumers happen to have metadata that don't conform to GeoDCAT-AP

So before we make a decision I think we should at least have a look at the existing mechanism in DCAT for metadata (catalogues)

If I understand it well it would be something like

:mydataset a dcat:Dataset;

:mydatasetRecordInCatalogueX a dcat:DatasetRecord;
  foaf:primaryTopic :myDataset
  dcterms:conformsTo :geoDCATAP.

This somehow assumes that there is a definition of the catalogue X as a dcat:Catalogue somewhere.
This is rather more complex, but as already hinted, I believe the original requirement makes sense if the data consumer looks at a specific piece of metadata, that's likely to be provided by a concrete dataset catalogue.

The solution has also the benefit that we don't have to sub-property dcterms:conformsTo with a new DQV property :-)



On 11/13/15 6:50 PM, Riccardo Albertoni wrote:
> Dear All,
> This mail follows  today's group call and the discussion on the use cases  DQV should  cover when dealing with "the compliancy of a dataset to a standard". (issue-202 [1])
>   Phil and Hadley suggested that the DQV should  express  assertions like
> "this dataset's metadata is conforming to a certain standards",
> and specifically to model this,  we discussed to add  dqv:hasMetadataConformingTo as  a new  subproperty of   dcterms:conformsTo.
> For example, DQV could  model the assertion
> " mydataset's metadata conforms geoDCAT-AP".
> with
> :mydataset a dcat:Dataset;
>   dqv:hasMetadataConformingTo  :geoDCATAP.
> :geoDCATAP a dct:Standard;
>    dct:title "GeoDCAT Application Profile developed in the context of the Interoperability Solutions for European Public Administrations (ISA) Programme"@en;
>    dct:issued "201X-XX-XX"^^xsd:date.
>   I'd like to thank Phil and Hadley for suggesting this new requirement.
> Is there any objection in adding this new kind of assertion in DQV expressivity?
> Do you have any comments or refinements on the modeling example?
> Have a good weekend!
> Riccardo
> [1]
> --
> ----------------------------------------------------------------------------
> Riccardo Albertoni
> Istituto per la Matematica Applicata e Tecnologie Informatiche "Enrico Magenes"
> Consiglio Nazionale delle Ricerche
> via de Marini 6 - 16149 GENOVA - ITALIA
> tel. +39-010-6475624 - fax +39-010-6475660
> e-mail: <>
> Skype: callto://riccardoalbertoni/
> LinkedIn:
> www:

Received on Friday, 27 November 2015 08:50:31 UTC