– DRAFT –
Dataset Exchange Working Group Teleconference

<riccardoAlbertoni> me hi dave, do you need the webex link?

<riccardoAlbertoni> PROPOSED: approve last meeting minutes https://www.w3.org/2021/03/03-dxwgdcat-minutes

Last minutes

<AndreaPerego> +1

+1

<riccardoAlbertoni> 0 i was not there

draft minutes v2

<riccardoAlbertoni> https://www.w3.org/2021/03/03-dxwgdcat-minutes

<alejandra> +1

Resolution: approve last meeting minutes https://www.w3.org/2021/03/03-dxwgdcat-minutes

<riccardoAlbertoni> https://www.w3.org/2017/dxwg/wiki/Meetings:Telecon2021.03.17

approve agenda

riccardoAlbertoni: Any updates to agenda?

<AndreaPerego> +1

riccardoAlbertoni: Silence indicates assent

<alejandra> I have a few questions about vocabularies that we adopt

Oustanding issues

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues?q=is%3Aissue+is%3Aopen+label%3A%22requires+discussion%22+label%3Afeedback

alejandra: eg checksum etc

<riccardoAlbertoni> Assign to volunteers the issues that can be advanced via draft PR Discuss the remaining in two hours SPRINT and then split among editors Deal with one/two of them at the beginning of each meeting

<riccardoAlbertoni> 1) Assign to volunteers the issues that can be advanced via draft PR 2) Discuss the remaining in two hours SPRINT and then split among editors 3 ) Deal with one/two of them at the beginning of each meeting

riccardoAlbertoni: Approach to organising work for outstanding issues. Assign to volunteers or use sprint?

riccardoAlbertoni: ...Any other ideas or suggestions?

<alejandra> +1 to Andrea's comments

AndreaPerego: Need to respond to some of these issues from outside the group - some of some age (2019).

AndreaPerego: .... they do deserve a reply/response

AndreaPerego: .... Some others could get a quick decision/reply so maybe during a quick call

riccardoAlbertoni: Do we want to go into a single issue now? Can list them in next agenda?

<riccardoAlbertoni> can we vote for 1+3 ?

<alejandra> +1

+1

<AndreaPerego> +1

<riccardoAlbertoni> +1

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1246

riccardoAlbertoni: I can handle this - is it controversial?

AndreaPerego: Well, some of it is specific...

AndreaPerego: ....e.g point 4

riccardoAlbertoni: Well, I was volunteering for parts 1+2 of that.....if it needs more specific expertise....?

AndreaPerego: Perhaps do a PR on 1+2 and then start conversation on rest?

AndreaPerego: .... let me document in an action....

Action: riccardoAlbertoni to create a PR to address points 1 and 2 of https://github.com/w3c/dxwg/issues/1246

<trackbot> Created ACTION-439 - Create a pr to address points 1 and 2 of https://github.com/w3c/dxwg/issues/1246 [on Riccardo Albertoni - due 2021-03-24].

Action: AndreaPerego to reply to points 3 and 4 of https://github.com/w3c/dxwg/issues/1246

<trackbot> Created ACTION-440 - Reply to points 3 and 4 of https://github.com/w3c/dxwg/issues/1246 [on Andrea Perego - due 2021-03-24].

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1126

riccardoAlbertoni: Are there others that can be advanced by PR?

riccardoAlbertoni: I will make sure we have a list in the next agenda

Feedback on PR #1323 about Distribution digest

<AndreaPerego> https://github.com/w3c/dxwg/issues/1323

<alejandra> This is the PR: https://github.com/w3c/dxwg/pull/1323

AndreaPerego: Essentially to allow confirmation that what a user gets from the distribution is what they expect (checksum)

alejandra: Agree this is necessary - important use case.

alejandra: ... Links I'd found for spdx were quite old but now see that there is something more modern in github

<alejandra> This is the information I had found before:

<alejandra> https://wiki.spdx.org/view/SPDX_FAQ

AndreaPerego: Standard way to cover licences etc with extensive list. Not implemented as rdf but quite good coverage

<alejandra> but I notice that Andrea pointed to https://spdx.dev/

AndreaPerego: ... seems to be stable/well maintained

<riccardoAlbertoni> ack

AndreaPerego: ... if we have concerns we could duplicate (like we did with PAV) but dont think is needed - this has active maintenance

riccardoAlbertoni: I think this looks good to me.

riccardoAlbertoni: ... there are a couple of questions still open...

riccardoAlbertoni: ... around algorithms.

AndreaPerego: ... we could probably defer that till later

AndreaPerego: ...not a great issue since spdx is broadly used so the capabilities should exist to support sensible behaviour

alejandra: That clarifies a lot

<alejandra> https://spdx.org/rdf/terms/

<alejandra> https://spdx.github.io/spdx-spec/

alejandra: is that the latest?

alejandra: More generally, do we have some rules about when we handle like we did with PAV or when we can rely on something existing directly?

<riccardoAlbertoni> Data on the Web Best109Practice #15 “Use terms from shared vocabularies, preferably standardized1104 ones, to encode data and metadata.”

AndreaPerego: Question is if there is adequate governance behind some standard/recommendation. Perhaps for Philippe

alejandra: Good - consideration of algorithms would be good

alejandra: ... Should make sure we point at the best references in the bibliography

<AndreaPerego> https://spdx.dev/specifications/

alejandra: Need a route through to RDF. Also we should check the licence....

alejandra: ... which looks okay

<AndreaPerego> https://github.com/w3c/dxwg/issues/1322

<riccardoAlbertoni> https://www.w3.org/TR/prov-o/#inverse-names

riccardoAlbertoni: Not the first inverse property request we have had

riccardoAlbertoni: Prov-O handled as per link - so in generally don't have too many inverses, but a way to create them in specific circumstances

AndreaPerego: Good approach - need to work through details

riccardoAlbertoni: We are constrained by properties we inherited in earlier DCAT versions
… need to double check

AndreaPerego: We may need to have two groups of properties - some with explicit inverse, some with a more implicit version

<AndreaPerego> proposed: let's follow the PROV approach for inverse properties

<AndreaPerego> +1

<alejandra> +1

<riccardoAlbertoni> +1

+1

Resolution: let's follow the PROV approach for inverse properties

<AndreaPerego> https://www.dublincore.org/specifications/dublin-core/dcmi-terms/#http://purl.org/dc/dcam/domainIncludes

<alejandra> https://github.com/w3c/dxwg/issues/912

AndreaPerego: Different vocabularies address different communities so we should take that into account when we decide to adopt something from elsewhere

<riccardoAlbertoni> See discussion in comment https://github.com/w3c/dxwg/issues/1307#issuecomment-800595335 and below.

riccardoAlbertoni: if people can review PR and get something through to ED.

AndreaPerego: Wanted to raise the issue since I was aware of it. We need to look at consistency we might need to adjust a little to make it practical.

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1307#issuecomment-800595335

riccardoAlbertoni: Continue this discussion in above thread since we have others interested

AndreaPerego: Will do

<AndreaPerego> Thanks, and bye