W3C

– DRAFT –
DXWG DCAT subgroup teleconference 19 December 2018 21:00 UTC

19 December 2018

Meeting minutes

<alejandra> rrsagent: make logs public

Confirm agenda

<alejandra> https://‌www.w3.org/‌2017/‌dxwg/‌wiki/‌Meetings:DCAT-Telecon2018.12.19

Resolved: agenda confirmed (noting that dataset relations and versioning on milestone

Approve minutes from last meeting

confirm minutes

<alejandra> https://‌www.w3.org/‌2018/‌12/‌12-dxwgdcat-minutes

<alejandra> +1

0 not present

<DaveBrowning> +1

<riccardoAlbertoni> +1

Resolved: minutes confirmed

open actions

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

DaveBrowning: some overlap with GH issues
… some tidying up possible

Action: DaveBrowning to mark actions closed where possible

<trackbot> Sorry, but no Tracker is associated with this channel.

actions outstanding - responses to Clemens and Luca

<alejandra> SimonCox: I was planning to reply along the lines that catalogues of services was a requirement

<alejandra> ... and that there are existing catalogues of services that need to be formalised

<alejandra> ... yes, this requires an extension of scope for DCAT

<alejandra> ... but we aren't being original, we are formalising and standardising what is being used

SimonCox: to work on draft today ... share with editors

<alejandra> alejandra: in agreement

DaveBrowning: is concerned about interactions between services, distributions and services
… we have not solved all the problems - some challenges ahead

DaveBrowning: partly motivated by conersations in Refinitiv
… versions of services, extracts of distributions

alejandra: please describe issues and add to GH asap

<alejandra> I think the issues already exist

<alejandra> SimonCox: I hear what you're saying, I suspect that the problem is more with the notion of versioning rather than the infrastructure to describe services

<alejandra> ... yes, there will be interactions, there is a need to formalise versioning

<alejandra> ... my concern is not to let the tail wag the dog

<alejandra> ... for many applications, versioning would be a minor concern

<DaveBrowning> +1 to Simon's perspective

<alejandra> ... we shouldn't get suck down to that vortex

<alejandra> ... with the treatment of services we have described what has already been done

<alejandra> Luca's comment: https://‌lists.w3.org/‌Archives/‌Public/‌public-dxwg-comments/‌2018Dec/‌0000.html

moving on to Luca's latest comment

alejandra: Andrea has proposed part of reply
… EPOS-DCAT-AP many more types are catalogued. Is this OK for DCAT scope - what subclasses of dcat:Resource should be considered?

<PWinstanley> s/catlaogued/catalogued/

riccardoAlbertoni: in Open World anyone can sub-class dcat:Resource!
… but am concerned about overextending scope of DCAT
… we should provide some examples and guidance
… noting that it is not a closed-world

alejandra: DCAT scope is fixed by requirements. We have nothing to say about extension of scope beyond that

alejandra: luca's 2nd point - composition of resources
… related to hasPart issues already on our list
… hasPart also relates to equipment
… so does hasPart apply to dcat:Resource? (and not just dcat:Dataset)

<alejandra> SimonCox: when we refactored the model to create the 'abstract' superclass dcat:Resource

<alejandra> ... the choice of promoting properties to dcat:Resource

<alejandra> ... was done according to the properties existing at the time

we should promote other properties to dcat:Resource if now requirements come to light

<alejandra> from Andrea's email: "About the "project context", Luca confirmed they have requirements here (but he just provided an example), and it may be worth asking him to verify whether they are or not covered by our use cases and, if not, to complement them."

<alejandra> alejandra: we still have issues related to this, e.g. related to Simon's project ontology

alejandra: prioritize outstanding issues based on external feedback?
… luca's final comment on Hydra - we should add an example based on this?

<alejandra> references to Hydra services: https://‌github.com/‌w3c/‌dxwg/‌issues/‌530

alejandra: is already mentioned in #530

<PWinstanley> https://‌www.hydra-cg.com/‌spec/‌latest/‌core/

Action: create GH issue to create Hydra example

<trackbot> Sorry, but no Tracker is associated with this channel.

<PWinstanley> isn't this dealt with by OWA?

alejandra: Andrea had open action to respond
… to Luca

<PWinstanley> do we need to have examples from every API spec?

<PWinstanley> so why add Hydra, specifically?

<alejandra> SimonCox: since we've moved all but the skeleton examples to an annex, I don't see a risk in adding as many examples as people are willing to do

<PWinstanley> ok, good point SimonCox

<alejandra> ... just a matter of obtaining them

<PWinstanley> it is the latter

<PWinstanley> OK !

<PWinstanley> +1

<alejandra> alejandra: potentially to be included in the primer

<alejandra> SimonCox: also in the repository

also in GH repository

Action: Andrea to respond to Luca

<trackbot> Sorry, but no Tracker is associated with this channel.

<alejandra> https://‌github.com/‌w3c/‌dxwg/‌milestone/‌14

milestone 3

https://‌github.com/‌w3c/‌dxwg/‌milestone/‌14

1. Riccardo's PR - on identifiers https://‌github.com/‌w3c/‌dxwg/‌pull/‌614

can be closed

<alejandra> Riccardo's PR can close two issues: https://‌github.com/‌w3c/‌dxwg/‌issues/‌53

<alejandra> https://‌github.com/‌w3c/‌dxwg/‌issues/‌68

alejandra: still has some ideas on alternative identifiers -
… has not added them to issue :-(
… yes, identifier patterns could/should be documented in Application Profiles

<alejandra> suggest keep this issue open: https://‌github.com/‌w3c/‌dxwg/‌issues/‌67

alejandra: and then discuss alejandra's proposals later

riccardoAlbertoni: thinks issues can be closed

riccardoAlbertoni: thinks Makx concerns about 'primary' 'secondary' etc belongs in an AP

alejandra: but there is no guidance currently on how to express this

alejandra: merge PR 614, close #53, #68

+1

<DaveBrowning> ok

<riccardoAlbertoni> +1

<PWinstanley> +1

Resolved: merge PR 614, close #53, #68

<alejandra> SimonCox: I worked on issue 79 because it was related to the qualified relations

<alejandra> ... very discrete things with potentially lots of power

<alejandra> Versioning text in the wiki: https://‌github.com/‌w3c/‌dxwg/‌wiki/‌Dataset-(and-other-DCAT)-versioning

DaveBrowning: versioning is stuck. Not sure how to progress - needs help
… probably will not be ready to the next PWD
… need to solve the problem, but currently blocked
… alejandra early contributions were good, now needs to be synthesized
… how much can we really say about versioning normatively

<alejandra> SimonCox: are we making this more complicated than it needs be? there is PAV

<riccardoAlbertoni> +1 pav as starting point, it is coherent with the DWBP https://‌www.w3.org/‌TR/‌dwbp/#dataVersioning

<alejandra> https://‌www.w3.org/‌TR/‌hcls-dataset/

alejandra: HCLS dataset description discussed versioning a lot, draws on PAV
… split version level and distribution level

DaveBrowning: agrees that PAV meets at least some of the requirements
… might unblock the matter
… needs some discursive material on versioning choices
… but perhaps this belongs in Primer not Rec

DaveBrowning: has some interesting examples that could be described in January - maybe don't belong in rec

alejandra: PAV is not a Rec. ... does this matter?

riccardoAlbertoni: recalls discussion on re-use of 3rd party vocabs in relation to DQV
… maintainable, persistently accessible is the main criterion, not formal status
… will it be viable long term?

<alejandra> alejandra: we might deal with sustainability of the resource by integrating the solution

persistence issue might be managed by publishing as a W3C Note ...

<alejandra> q/

<alejandra> ?

riccardoAlbertoni: did not use vocabularies where the definitions were not clear or mods were needed

<alejandra> https://‌github.com/‌w3c/‌dxwg/‌issues/‌248

assigning owners to issues in milestone

alejandra: License and Rights - need to add some properties into text description of DCAT classes

DaveBrowning: agrees

<PWinstanley> How well maintained is PAV?

<PWinstanley> according to LOV the last version was c.4 y ago

<PWinstanley> ok, that's helpful.

<riccardoAlbertoni> s/riccardoAlbertoni: did not use vocabularies where the definitions were not clear of mods were /riccardoAlbertoni: In DQV we have ingested daQ, as we needed to make it more compliant with the DWBP group expectation

PAV is a profile of PROV ... !

could be documented using the Profiles ontology

Work needed on document:

alejandra: to copy properties into classes

DaveBrowning: concerned about contributor list - header or annex?

alejandra: depends on length

DaveBrowning: contributors to meetings and PRs
… should be included

<alejandra> https://‌github.com/‌w3c/‌dxwg/‌milestone/‌14

alejandra: we can close most of the milestone issues
… can riccardoAlbertoni comment on FAIR?

riccardoAlbertoni: move to next milestone
… maybe Andrea has specific suggestion

DaveBrowning: Andrea made 'strong recommendation' to address this in the currnet milestone

<alejandra> https://‌github.com/‌w3c/‌dxwg/‌milestone/‌15

DaveBrowning: next MS!

Xmas homework

Next meeting in week of 7th Jan

<riccardoAlbertoni> happy end of the year

<PWinstanley> bye!! Have a great holiday. Happy Christmas

<alejandra> happy holidays!

Summary of action items

  1. DaveBrowning to mark actions closed where possible
  2. create GH issue to create Hydra example
  3. Andrea to respond to Luca

Summary of resolutions

  1. agenda confirmed (noting that dataset relations and versioning on milestone
  2. minutes confirmed
  3. merge PR 614, close #53, #68
Minutes manually created (not a transcript), formatted by Bert Bos's scribe.perl version 2.49 (2018/09/19 15:29:32), a reimplementation of David Booth's scribe.perl. See CVS log.

Diagnostics

Succeeded: s/services/treatment of services/

Succeeded: s/catlaogued/catalogued/

Failed: s/catlaogued/catalogued/

Succeeded: s/ANdrea/Andrea/

Succeeded: s/beyind/beyond/

Succeeded: s/?q/

Succeeded: s/q//

Succeeded: s/of/or/

Failed: s/riccardoAlbertoni: did not use vocabularies where the definitions were not clear of mods were /riccardoAlbertoni: In DQV we have ingested daQ, as we needed to make it more compliant with the DWBP group expectation