– DRAFT –
DXWG DCAT subgroup teleconference 31 July 2019

approve agenda

<DaveBrowning> https://‌www.w3.org/‌2017/‌dxwg/‌wiki/‌Meetings:DCAT-Telecon2019.07.31

DaveBrowning: That's pretty simple. Any suggests changes / additions?

PWinstanley: We could also include the question I made to Philippe - whether or not in non-normative sections we can refer to document which are still draft.

PLH: Not a problem at all.

<DaveBrowning> Minutes of last meeting : https://‌www.w3.org/‌2019/‌07/‌24-dxwgdcat-minutes

approve last meeting minutes

<SimonCox> +1

<DaveBrowning> +1

<PWinstanley> +1

<alejandra> +1

+1 modulo add me please

Resolved: last meeting minutes approved

summary for plh

DaveBrowning: We published a new WD and proactively asked feedback . We got a good amount.
… We basically incorporated most of the comments.
… There are still open issues, but they are close to be solved.
… We can see the end of it, although there is still some work to be done.
… Some of them are about the .ttl document.
… So, we should be able to complete the document shortly.
… We have however 2 questions:
… (1) mechanics about RDF implementation
… (2) timing to implementation report.
… Any comments / questions?

PWinstanley: BTW, we had also a poll about moving DCAT to CR.
… This is settled [12 positive responses out of 12]

plh: So, it seems there are not substantive changes, only editorial, right?

DaveBrowning: Yep.

plh: This then should not block the steps to the director to move to CR.
… So, you just have to say that you are in the process of addressing all the comments.
… Did you get comments from the privacy people?

PWinstanley: I made a presentation to the privacy people, but no substantive feedback.

plh: So this should be settled.

PWinstanley: What they said is that if we do a primer, then that could be the place to re-enforce the message of issues.
… So they see privacy issues more in scope in a document about implementation of DCAT rather than in the DCAT spec itself.

alejandra: I wonder what may happen in case we get substantive comments while already in the CR process.

plh: That would be unfortunate. You risk to have a delay up to 30 days.
… But I suggest you go ahead and you deal with this if this happen.

<alejandra> ok, thanks

plh: Anyway, in such a case, you have to publish a new CR, but you have just to explain what is different.
… There's also the call for exclusion.

DaveBrowning: So, coming back to our questions...

rdf files

<PWinstanley> https://‌github.com/‌w3c/‌dxwg/‌pull/‌881#issuecomment-516859294

DaveBrowning: The new spec uses the same namespace of the original one.
… The current namespace page of DCAT points to different serialisations of the 2014 DCAT voc.
… So, our idea is to add some more links, but we replace the current files with the ones with the new version.
… We need some guidance on how to do this.

plh: It is up to you to decide.
… But we can link only to 1 .ttl file for the namespace URI.
… You can then use a different URL for the old one.

SimonCox: Another issue is that the link from DCAT 2014 will point to DCAT 2.

AndreaPerego: I am not aware of DCAT implementations which owl:import the vocabulary, but there may be.
… I wonder whether, plh, if W3C is thinking about version URLs for vocabularies as done for specs.

plh: The person you could ask is Ralph Swick so you can ask feedback on what you want to do.
… Is this a blocking issue?

DaveBrowning: I don't think so, as it is not normative.

plh: So you can go ahead, while asking feedback to Ralph.

Action: DaveBrowning to contact Ralph Swick about guidance on vocabulary versioning

<trackbot> Created ACTION-354 - Contact ralph swick about guidance on vocabulary versioning [on David Browning - due 2019-08-07].

DaveBrowning: Then if everyone is happy we can move to the next agenda item

implementation report

DaveBrowning: We would like to check what we have to do, also considering we are in August.

plh: What you need to do for CR is to explain what we want to demonstrate.
… This may start a negotiation.
… Depending on whether you have or not implementation evidence at the time of transition to CR, you can already move to the next step.

plh: Do you want me to send everyone the template to be filled in?

PWinstanley: Yes, please.

plh: So after your request, the director will take 7 days for publishing the CR.
… It is also possible to make some editorial changes before going out.
… Any substantive changes will require publishing a new CR.
… BTW, we want to be sure you are not touching any patent.
… Noting also the that notion of "substantive" may be subjective.

<PWinstanley> can we get a legal read-through (to cover most of the patent concern)?

<Zakim> PWinstanley, you wanted to ask can we get a legal read-through (to cover most of the patent concern)?

plh: About substantive comments coming in during CR, this is not so frequent.

PWinstanley: Can anyone at W3C have a check to see if anything in DCAT may trigger licences?

plh: Absolutely not.
… About legal issues, this happens some times, but we address this with the calls for exclusions.
… The issue is when the relevant companies are not in WG, so they miss it.

DaveBrowning: So it seems we need to have our implementation criteria properly described.

plh: Can someone describe on the fly what this will be?

<DaveBrowning> AndreaPerego: The new DCAt extends the original version. SO the idea is to focus on changes

<DaveBrowning> ... Have started looking at work done with profiling of DCAT 2014 as examples of what we've introduced

<DaveBrowning> ... Other area is catalog platforms e.g. CKAN or existing catalogs for other uses

<DaveBrowning> ... we know that people have been extending DCAT 2014 to address the same needs as we have in DCAT 2

<DaveBrowning> ... so new version addresses the gaps in DCAT 2014

<plh> https://‌www.w3.org/‌2019/‌Process-20190301/#implementation-experience

<DaveBrowning> PWinstanley: There is also discussion in various groups to adopt DCAT 2

plh: E.g., within the EU ISA2 programme there is an explicit reference to the new version of DCAT, considering to take into account in future revisions of their vocabularies.

plh: One of the questions is whether you have any independent implementation...
… Focussing on the changes only is fine.
… About other documents in the pipeline, as the Verifiable Claims group, they can provide some example on how to formulate the implementation evidence criteria.

<plh> https://‌www.w3.org/‌TR/‌2019/‌CR-vc-data-model-20190725/

<plh> "The Working Group has received implementation feedback showing that there are at least two implementations for each normative feature in the specification."

SimonCox: If we find features in the spec that do not have 2 implementations, should we remove them, make them non-normative, ...

plh: My advise is to mark them as features at risk. This way you can remove them without publishing a new CR.
… Otherwise, removing them will be considered a substantive change.

<alejandra> https://‌www.w3.org/‌TR/‌2019/‌CR-vc-data-model-20190725/#json-web-token

plh: The document I pointed you to has a section marked as at risk.

SimonCox: So the spatial and qualified relationships can be marked as at risk.

<SimonCox> perhaps we should mark the Location and QualifiedRelation sections, maybe some of the DataService aspects 'at risk'

DaveBrowning: So, they will be either removed or kept.

plh: They could also be made non-normative.

SimonCox: We should create an issue about marking features at risk.

DaveBrowning: I'll do that.

[meeting adjourned]

s/PWinstanley: E.g., within the/plh: E.g., within the/

s/plh: E.g., within the/PWinstanley: E.g., within the/

s/PWinstanley: E.g., within the//

s/PWinstanley: E.g., within the//

s|s/PWinstanley: E.g., within the/plh: E.g., within the/||

s|s/PWinstanley: E.g., within the/plh: E.g., within the/||