DCAT sub-group 2018-05-31

31 May 2018

Meeting Minutes

<SimonCox> Webex https://‌csiro.webex.com/‌csiro/‌j.php?MTID=m90c9cd8d1e6b406fe4b68e5cf0c09d78

confirm agenda

<SimonCox> insert briefing from from roba and nick

Approve minutes from last meeting – https://‌www.w3.org/‌2018/‌05/‌24-dxwgdcat-minutes

<AndreaPerego> +1

<SimonCox> +1

<PWinstanley> 0 - not there

<riccardoAlbertoni> +1

<roba> +1

0 (missed it, conflicting meeting)

Resolved: minutes of last meeting approved

Need a volunteer to do analysis of national DCAT-AP profiles to verify that all added classes are on our agenda https://‌lists.w3.org/‌Archives/‌Public/‌public-dxwg-wg/‌2018May/‌0401.html

AndreaPerego: Max was in charge of the work of an analysis of the adoption of DCAT-AP
… i can help as well, but it depends on how much work that would entail

Action: AndreaPerego to check if Makx is available

roba: At yesterday's plenary meeting I prepared a use case around DCAT-AP

<AndreaPerego> About the report on the use of DCAT-AP (that may be used as a starting point for a comparative analysis of national implementations): https://‌joinup.ec.europa.eu/‌document/‌report-dcat-ap-use

roba: glaring obvious omission in the use cases and requirements IMHO
… Phil Archer pointed to Stijn as a potential contributor

<Jaroslav_Pullmann> I am sorry for being late!

<Stijn_Goedertier_AIV> happy to contribute to the analysis

roba & nick to brief on profiledesc work

roba: Developed a new alignment file
… matches the example of best practice of cataloging that Max gave in the F2F

<SimonCox> See https://‌github.com/‌w3c/‌dxwg/‌blob/‌gh-pages/‌profiledesc/‌profiledesc_dcat_alignment.ttl

<AndreaPerego> Again about DCAT-AP national extensions, this study could be even more relevant: https://‌joinup.ec.europa.eu/‌document/‌national-extensions-analysis-dcat-ap

roba: also made some small changes the profiledesc vocabulary

<SimonCox> required some tweaks to https://‌github.com/‌w3c/‌dxwg/‌blob/‌gh-pages/‌profiledesc/‌profiledesc.ttl

<AndreaPerego> ... and maybe this one as well: https://‌datos.gob.es/‌sites/‌default/‌files/‌doc/‌file/‌report_dcat-ap_and_its_extensions.pdf

roba: Nick did some work on updating the examples and put in some normative resource pointers
… request the DCAT-AP experts to validate the examples

<SimonCox> Examples https://‌github.com/‌w3c/‌dxwg/‌tree/‌gh-pages/‌profiledesc/‌examples

<SimonCox> https://‌github.com/‌w3c/‌dxwg/‌blob/‌gh-pages/‌profiledesc/‌examples/‌profiledesc_examples.ttl

<SimonCox> and https://‌github.com/‌w3c/‌dxwg/‌blob/‌gh-pages/‌profiledesc/‌examples/‌eg_dcat-ap2.png

<PWinstanley> https://‌ec.europa.eu/‌isa2/‌solutions/‌dcat-application-profile-data-portals-europe_en

<roba> https://‌github.com/‌w3c/‌dxwg/‌blob/‌gh-pages/‌profiledesc/‌examples/‌profiledesc_examples.ttl

<SimonCox> in particular check these URIs <https://‌joinup.ec.europa.eu/‌rdf_entity/‌http_e_f_fdata_ceuropa_ceu_fw21_f6f27f059_bf785_b4d7d_bb602_b6448aab73bd5>

<AndreaPerego> I can send you the "official" URIs for DCAT-AP etc.

SimonCox: You used a subclass from the new dcat:resource

Nick: you will be able to get distributions through content negotiations in this proposal

<AndreaPerego> https://‌www.w3.org/‌TR/‌vocab-dcat-2/#bib-DCAT-AP

AndreaPerego: The URIs are included in the reference

<SimonCox> look at https://‌www.specref.org/

<AndreaPerego> https://‌www.specref.org/?q=dcat-ap

<PWinstanley> https://‌www.difi.no/‌fagomrader-og-tjenester/‌digitalisering-og-samordning/‌standarder/‌standarder/‌dcatdata-catalog-vocabulary

<PWinstanley> https://‌doc.difi.no/‌dcat-ap-no/

<AndreaPerego> Namespaces for DCAT-AP etc.: http://‌data.europa.eu/‌URI.html

<AndreaPerego> GeoDCAT-AP - data.europa.eu/930/

<AndreaPerego> DCAT-AP - data.europa.eu/r5r/

Action: AndreaPerego to assist verification of the URIs in the DCAT-AP profiledesc examples

<NicholasCar> Here're the examples in HTML

<NicholasCar> https://‌w3c.github.io/‌dxwg/‌profiledesc/‌examples/

AndreaPerego: Is there are relationships between different versions of DCAT-AP's?

<PWinstanley> https://‌github.com/‌difi/‌dcat-ap-no/‌issues

roba: Versioning in general would be a different type of relationship. dcat:revision to be backward compatible, i.e. original DCAT would become a profile of the new DCAT

AndreaPerego: I am not sure if these two relations can be modelled with the same relationship type.

roba: Suggest to remove the versioning example.

SimonCox: All of the conversation was about the relationships between versioning and resources.
… most of that will be covered in the profile work item.

Discussion of priorities and actions

<roba> * Nick - we should split the version example into a separate file, with a more exploratory status...

SimonCox: Most of the discussion on the mailing list recently has been on the profile discussion, rather than on DCAT itself

<SimonCox> maybe look here https://‌github.com/‌w3c/‌dxwg/‌milestones

SimonCox: This should give us some view on what areas need some work.

riccardoAlbertoni: I have added an milestone on Quality in DCAT collecting two issues.

<SimonCox> suggestion: relationships between resources (including datasets, profiles)

<SimonCox> related to versioning

SimonCox: I know that you nick were interested in qualified relationships in general.

<SimonCox> related to partonomy

NicholasCar: I am keen to do some work with qualified relations.
… introduce some modelling patterns from some other W3C working groups/

<SimonCox> qualified forms works for agent relations, dataset relationships, versioning relationships, using code-list for role of relationship

<AndreaPerego> I included some examples here: https://‌github.com/‌w3c/‌dxwg/‌wiki/‌Provenance-patterns

AndreaPerego: I added some examples on qualified and non-qualified forms on the wiki page
… they are taken from the use cases
… we ended up using the PROV ontology, and we validated the use of PROV with the working group members
… they did not use a specific relation that is in PROV, i.e. hadRole themselves
… so it would be worth checking with the PROV people on the examples

SimonCox: Are these qualified relations changing the DCAT vocabulary, or are they just best practices how to use the vocabulary in combination with other vocabularies

AndreaPerego: Yes, we don't need to add anything. We just recommend how properties can be used in qualified relations using the PROV ontology

<SimonCox> ... look at PROV mappings to DC as exemplars?

AndreaPerego: we have already a good starting point from the PROV working group where they proposed examples on how to use Dublin Core with PROV
… just guidance on how to use the PROV ontology to qualify resources

NicholasCar: The examples with agent role is just an easy one, but the real one would be around Datasets and Distributions

<SimonCox> NicholasCar: real target of qualified form is dataset-dataset relations, ...

NicholasCar: we have done some thinking in the RDA working group how to do that, i.e. model the Dataset as a prov:entity

<SimonCox> ... dataset a prov:Entity !

AndreaPerego: Another example besides roles, are use cases around the funding source, i.e. context of the dataset creation
… especially with scientific datasets

<SimonCox> 1. agent roles

<SimonCox> 2. dataset-relations

<SimonCox> 3. dataset context

<AndreaPerego> +1

<NicholasCar> yes

roba: what is the method to hand this over to the guidance group?

<riccardoAlbertoni> +1 to have more guidance in dcat document

SimonCox: My assumption is that the DCAT document needs to include more documentation, guidance and examples.
… expectation is that we have more example in the document.

<AndreaPerego> Indeed, the lack of examples in DCAT 1.0 has always been an issue.

SimonCox: preference is that the document becomes more guiding rather than just a spec
… however, guidance around application profiles will be in other documents rather than in the DCAT spec

<roba> +1

<NicholasCar> +1

<NicholasCar> no

SimonCox: All PROV relations are dealing with change. They do not include static relations.

NicholasCar: I have not looked into non-PROV qualified relations that we can include

<SimonCox> qualifiedAttribution can support static relations!

NicholasCar: qualified attribution pattern is something that we need to include
… and can include static relations

<AndreaPerego> I can help for sure, but don't have time for being in charge of it.

<NicholasCar> seriously injured child, must go

<AndreaPerego> Sorry to hear that, Nick.

Action: SimonCox to coordinate qualified relations topics

AndreaPerego: Qualified relations will also help with the quality modelling

<riccardoAlbertoni> yes

<riccardoAlbertoni> it is probably related to issue 58

<roba> bye

<riccardoAlbertoni> byee

<Stijn_Goedertier_AIV> thanks all

<Jaroslav_Pullmann> bye!

<AndreaPerego> [meeting adjourned]

Summary of Action Items

  1. AndreaPerego to check if Makx is available
  2. AndreaPerego to assist verification of the URIs in the DCAT-AP profiledesc examples
  3. SimonCox to coordinate qualified relations topics

Summary of Resolutions

  1. minutes of last meeting approved
Minutes formatted by Bert Bos's scribe.perl version 2.41 (2018/03/23 13:13:49), a reimplementation of David Booth's scribe.perl. See CVS log.


Succeeded: s/DCAT/dcat

Succeeded: s/to be a/to be

Succeeded: s/I have added an issue on Quality in DCAT./I have added an milestone on Quality in DCAT collecting two issues.

Succeeded: s/rrsagent draft minutes v2//