– DRAFT –
DXWG DCAT subgroup call

approve last meeting minutes

<riccardoAlbertoni> PROPOSED: approve last meeting minutes https://www.w3.org/2020/11/11-dxwgdcat-minutes

<Makx_> 0

+1

<riccardoAlbertoni> +1

Resolution: approve last meeting minutes https://www.w3.org/2020/11/11-dxwgdcat-minutes

approving agenda

<riccardoAlbertoni> https://www.w3.org/2017/dxwg/wiki/Meetings:Telecon2021.01.20

AndreaPerego: Fine with me

<riccardoAlbertoni> https://github.com/w3c/dxwg/projects/17

open issue on gh

riccardoAlbertoni: Issues have been grouped for being reviewed by topic.
… We have a lot of stale issues.
… Anyway, when closing issues, especially those concerning requirements, we should make a statement there.
… Should we start with them?

[silence]

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/82 Reference system [RRS]

riccardoAlbertoni: Let's stat then with them.

<riccardoAlbertoni> https://www.w3.org/TR/vocab-dcat-2/#ex-dataset-crs

riccardoAlbertoni: For #82, the requirements should have been addressed in DCAT2 by providing an example of how to specify a CRS.

<riccardoAlbertoni> PROPOSED: close https://github.com/w3c/dxwg/issues/82

<Makx_> +1

<riccardoAlbertoni> +1

+1

<PWinstanley> +1

Resolution: close https://github.com/w3c/dxwg/issues/82

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/70 Publication control [RPC]

riccardoAlbertoni: Let's look at #70.

<riccardoAlbertoni> https://w3c.github.io/dxwg/dcat/#license-rights

riccardoAlbertoni: We do have in DCAT2 a section on licenses and rights.
… Either we keep things as they are, or we state that the same rules apply to dcat:CatalogRecord's

Makx: I don't see in that section about the type of resources a license applies to.

riccardoAlbertoni: It is implictly related to subclasses of dcat:Resource, which a dcat:CatalogRecord is not.
… It is however true that we have not specified the properties for all classes, and since we are in OWA one can use rights on any resource.

<Makx_> +1 to keep things as they are

AndreaPerego: I would keep things as they are. It was a requirement about IPR on metadata produced by private companies, and we didn't have any additional input along those lines.

riccardoAlbertoni: So we can just repeat what we said here.
… Any objections?

<Makx_> no objection

<PWinstanley> none from me

<riccardoAlbertoni> proposed: Reply to the issue explaining the open word assumpion and how to address the specific requirement without changing the document

<Makx_> +

+1

<Makx_> +1

<riccardoAlbertoni> +1

Resolution: Reply to the issue explaining the open word assumpion and how to address the specific requirement without changing the document

riccardoAlbertoni: Next one...

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1287 digest for DCAT distribution

riccardoAlbertoni: This is quite recent.
… There is a solution in DCAT-AP using spdx:checksum.
… My question is whether SPDX is stable enough to be included into a W3C Recommendation.
… WDYT?

Makx: Not sure on how much is stable.
… One of the issues raised is that there's only one way of expressing it.
… I think we should first check whether this solution is flexible enough.

<Makx_> Look at https://spdx.org/rdf/terms/#d4e1968

Makx: There was an issue about the support of only one hash algorithm, but the current version of SPDX supports more.

AndreaPerego: We need anyway to add some notes of use to make it clear when and for which use cases it can be used.

riccardoAlbertoni: Maybe we can start with a PR, and then work on that. Or we go on discussing on GH.
… WDYT?

AndreaPerego: I would propose to go on working on GH.

<Makx_> ok

riccardoAlbertoni: Any other comments?

riccardoAlbertoni: Next ones...

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/87 https://github.com/w3c/dxwg/issues/88

riccardoAlbertoni: About relationships with other vocs, not sure mapping with VoID is necessary.
… But in GH people are raising some questions on this.
… Should we elaborate more on this?
… We said that DCAT, VoID, QB can be used in a complementary way, we have DataID as an example.

Makx: I would just close it.

PWinstanley: Agreed.

<Makx_> No further elaboration

riccardoAlbertoni: My concern is that in #88 there was quite a long discussion on this.
… Some people were asking clarifications.
… After all, I agree we can close, at least the part about the mapping.
… On the other side, the comments were coming from people actually using DCAT.

AndreaPerego: I don't think we have enough evidence and use cases to understand how to link DCAT with VoID and QB.

riccardoAlbertoni: I actually don't think it is possible to standardise this link.

Makx: This is also what Bert was saying. And you also may need link DCAT from VoID.
… And not having use cases, we cannot provide a safe solution.

<riccardoAlbertoni> proposed: Closing the issues as we do not have enough use cases grounding a solution

<PWinstanley> +1

+1

<Makx_> +1

<riccardoAlbertoni> +1

<Makx_> +1

Resolution: Closing the issues as we do not have enough use cases grounding a solution

outstanding actions

<riccardoAlbertoni> https://www.w3.org/2017/dxwg/track/actions/open

riccardoAlbertoni: About https://www.w3.org/2017/dxwg/track/actions/417 , do you have anything, PWinstanley?

PWinstanley: I suggest we keep it, and collect the presentations we made to start a draft for the primer.

Makx: Should we have then a document linking to training material?

PWinstanley: Yes, plus some narrative.

riccardoAlbertoni: There is also a collection of resources I created, but I don't remember where it is.

<riccardoAlbertoni> https://github.com/w3c/dxwg/wiki/DCAT-presentations-and-external-materials

riccardoAlbertoni: Here ^^ !
… Anyone can add links to their presentations etc.

PWinstanley: Let's start this in April, when we have finished with the dissemination activities.

riccardoAlbertoni: About https://www.w3.org/2017/dxwg/track/actions/434 , do you have anything AndreaPerego?

AndreaPerego: It's a work item for the next drafts.

riccardoAlbertoni: This maybe relates to some feedback on versioning etc.

AndreaPerego: The action is about 2 aspects: providing more example and provide the mappings between unqualified and qualified relationships.
… About the examples, we need to decide the use cases we would like to use as a basis for the examples.
… About the mapping, we need to present them properly, as qualified relationships may rely on community-specific vocabularies (e.g., for roles) that would not ensure interoperability across communities and domains.

PWinstanley: Why not considering the use of RDF*, or property graphs, instead of PROV?

riccardoAlbertoni: So, should we keep the action, or make it an issue?

<Makx_> +1 to consider later

AndreaPerego: I think this maps to more that one issue, so I would keep it as an action, and decide which issues we should create in future meetings.

<riccardoAlbertoni> ack

<Makx_> nothing from me

AndreaPerego: Maybe, it would be worth knowing how much effort each of us can put in DCAT in the next few months, and decide what we can actually do for the next drafts.

Makx: An option is to set a date, and see what we can do.

PWinstanley: There is also the DCAT paper to be finalised and submitted.

riccardoAlbertoni: Let's meet in 2 weeks.

[meeting adjourned]