– DRAFT –
DCAT Plenary

https://www.w3.org/2022/07/12-dxwg-minutes

Proposed: to accept minutes https://www.w3.org/2022/07/12-dxwg-minutes

<riccardoAlbertoni> +1

<annette_g> +1

<Nobu_Ogura> +1

<pwinstanley> -0 not there

<pchampin> +1

+0

RESOLUTION: minutes accepted https://www.w3.org/2022/09/27-dxwg-minutes.html

TPAC update

Pierre-Antoine: my impressions. This group didn't have a meeting but the "data activity" was well represented
… I gave a lightening talk in the AC meeting around what's new
… Two new WGs: RDF Datasets and RDF-star
… Quite some discussion around JSON-LD: the WG is becoming active again
… Some discussion on YAML-LD
… Simple and yet more expressive than JSON. Makes some things easier

pwinstanley: anybody else from our WG attended?

annette_g: I did. I enjoyed the sustainability. Not direclty relevant for this group but interesting in general
… what's the impact of using web technology
… There was good stuff in the dev meetup. Queries/containers.

pwinstanley: what sort of scale is "green-computing" computing?

annette_g: not a lot of quantification yet.

<annette_g> https://almanac.httparchive.org/en/2022/sustainability

annette_g: some pointers were sent on W3C slack

pwinstanley: negotiation could be related to minimizing consumption
… if one does a lot of work with data there could be different versions.
… a very wild expansion of the notion of profiles
… sthg could have a lower energy utilization.
… This could be considered next to other concerns like ETL processing.
… We need to be thinking of all the things we do add up.

pwinstanley: is there sthg that hints that our WG should have contributed more?

riccardoAlbertoni: to me it was lack of energy. Other things had to be prioritized

Pierre-Antoine: it's hard to say
… DCAT is in good shape
… profile negotiation still feels rather early
… but it's going to take more space in the future
… I hope we can do something around RDF-star
… for example "I'd like Turtle but without the fancy RDF-star thing"
… We want to avoid being disruptive and creating a new content/media type for RDF
… We're in between things. So no lost opportunity but some more discussions

pwinstanley: cooperation with European Union people?

Pierre-Antoine: some have vocabularies published eg. on w3.org.
… maybe we'll have to find a home for this new work.

pwinstanley: a revitalization of the Gov WG?

annette_g: should we focused on DCAT and profile?

pwinstanley: I feel stronger about CAT than D :-)

annette_g: I feel the other way round :-)

pwinstanley: FAIR

antoine: which vocabularies?

Pierre-Antoine: Core vocabularies for persons, orgs...
… There are other people working on vocabularies. Core assessment methodologies.
… three groups of people have contacted me about similar things

pwinstanley: Nobu would there be some appetite from Japanese govt?

Nobu_Ogura: I would say yes

pwinstanley: W3C would give a neutral place for such work
… the "optics" would be easier
… there were also suggestions from US govt

Antoine: I saw sthg about evaluation methodologies on the EU portal for datasets

<pchampin> https://www.dagstuhl.de/en/program/calendar/evhp/?semnr=21383

Antoine: they may have idneed desires to standardize evaluation aspects
… but this would lead us into the "profile" work that may we were reluctant about several years ago.

Antoine: have there been discussions about "datasheets for datasets" ? This could be relevant for DCAT

pchampin: no

Pierre-Antoine: there was a meeting (25) organized by DDI and CODATA
… Some people use DCAT. Some Schema.org and DCAT perceived as too complicated

Pierre-Antoine: we talked about FAIR a lot.
… It's hard to turn the letters into technical measures
… Maybe this was the most obvious "missed opportunity"

pwinstanley: there's a week of activities in Leiden at the end of October
… I'll probably be there for some part
… The FAIR convergence symposium is one of the meetings.
… I'll send the link

<pwinstanley> https://codata.org/2nd-fair-convergence-symposium-24-26-october-2022-in-leiden-the-netherlands/

DCAT

riccardoAlbertoni: two meetings after the summer
… How to reply about the feedback got from TAG and Privacy group

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1530

<pwinstanley> https://codata.org/save-the-dates-hotzone-week-on-fair-at-the-2022-leiden-european-city-of-science-24-28-october/

<riccardoAlbertoni> https://github.com/w3c/dxwg/blob/gh-pages/docs/explainer.md.

riccardoAlbertoni: a first part of the answer is to add things to the explainer
… adding to the main features of DCAT 3 and the history.
… this part does not seem most critical. We suppose it will satisfy their curiosity
… The second element is why we've not used RDF lists for series
… dcat:first etc sounds like it is related to rdf:first etc

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1530#issuecomment-1255206996

riccardoAlbertoni: but they do have different semantics
… We explained that RDF lists based on LISP-style lists are not the best way to describe our datasets series
… Pierre-Antoine, Dave and XX alreayd say they agree
… Further feedback may be desired
… Any feedback?

PW

pwinstanley: I need to get back to the full presentation of DCAT v3
… we need better tutorials of how people are working with it

riccardoAlbertoni: my doubt is how to reply to such feedback
… what I've sent so far is clearly my position.

Pierre-Antoine: no clear advice. The answer was satisfying to me.

riccardoAlbertoni: when should the issue be closed?

Pierre-Antoine: I've not been in enough horizontal reviews. I would simply ask the question.

riccardoAlbertoni: ok

pwinstanley: sometimes things look too abstract, too distant to the domain

riccardoAlbertoni: I've tried to map the current datasets series to RDF lists, but it was hard
… I don't think people consuming dataset series function in LISP-list ways
… Maybe it was genuine curiosity

riccardoAlbertoni: let's see if they reply next week

Pierre-Antoine: I think it was to check that the group chose this option knowlingly.

pwinstanley: there was the PAV stuff brought in as well, for versioning

riccardoAlbertoni: one of our efforts was not to reivent the wheel

pwinstanley: there were reasons

riccardoAlbertoni: a long discussion about this re-use/import of PAV
… we mixed PAV with DCTerms

pwinstanley: I don't remember discussions around core RDF constructs.

riccardoAlbertoni: lists became more important than in the past.
… many people tried to stay away from RDF lists
… there was a paper about RDF lists, saying that they were not so well supported.
… The other thing was feedback from the privacy group

<riccardoAlbertoni> Discussion of the feedback from the privacy group https://github.com/w3c/dxwg/issues/created_by/npdoty

Antoine: cheap advice: ask them if they can live with your answer. People are never fully satisfied
… :-)

Also on RDF lists, if they have more questions maybe we could check what happened with Activity Streams.
… AS have come with their own "first" etc to handle pagination
… https://www.w3.org/TR/activitystreams-vocabulary/#dfn-first

<Zakim> annette_g, you wanted to mention web share API mention of DCAT -- typo?

<annette_g> https://www.w3.org/TR/2022/CR-web-share-20220830/

riccardoAlbertoni: it seems that the Privacy feedback is not centered on DCAT
… Please have a look at it

annette_g: the Web Share rec mentions DCAT

riccardoAlbertoni: I was not able to figure out what's happening

annette_g: it seems nobody knows?

annette_g: it's there