– DRAFT –
RDF-star WG

<pchampin> /www.w3.org/2023/12/22-rdf-star-minutes.html///www.w3.org/2022/12/22-rdf-star-minutes.html

<ora> Looking for an RDF-star meeting... ;-)

list attendees

ora: I have two agenda items: Naming thing, email exchange on the list

<Timothe> Unfortunatelly I have to leave. Is present- a thing?

Timothe: yes that should work

<Timothe> thanks, and sorry about that

ktk: there is a problem with IRC, we will do our best to scribe but several people cannot join IRC right now
… zoom will be used for queing for the moment (raise hand)

ora: Let's start with naming as first agenda item

naming of the specs

ora: two weeks ago the proposal was not quite ready.

pchampin: I did not realize there is a need to adjust the proposal

gkellogg: proposal can be found in issue https://github.com/w3c/rdf-star-wg/issues/4

AndyS: I like the remaining on some of the documents. there seems to be a consensus on the list that we will give advice on modeling with RDF Star. We might need a document for that and it probably should not be a recommendation.

ora: I agree this should be a non-normative document.
… we might form a task group to create an initial draft for that

ktk: couldn't the primer document be the place for such best practices?

pchampin: I think the 2004 version was a recommendation and it was made a "note" for the 1.1 version

ktk: we also mentionned that we wanted an equivalent to the sparql-overview document for the whole stack
… would this be a new document? integrated in sparql-overview?
… My personal preference would be to not add more documents. So recycling the sparql-overview for this.

pchampin: +1 on not having too many documents. I would have expected to have RDF primer as the overview

gkellogg: I agree we should not have too many documents. It could make sense to keep the SPARQL overview as the communities are not necessary the same people

<pchampin> example: https://www.w3.org/TR/swbp-n-aryRelations/

gkellogg: it does make sense to have something like RDF primer that can give some modelling support.

tallted: any individual document should not be overloaded. I think SPARQL is unique enough that we keep the overview doucument.
… we might rename SPARQL-overview to SPARQL-concepts

ora: We can re-visit the structure at a later date when we think we got it wrong
… I agree with what Ted said about not overloading documents too much
… the goal should be that there is a clear logical structure for all the documents we have. And we need a starting point for all of these
… after that it's the web, we can link. but there is a place where people can land and find all the related documents
… I propose that with some naming changes (if we need them) we proceed with the structure in the issue
… later on we can decide if the SPARQL-ovierview is/should be part of another document
… renaming it to SPARQL-concepts" is a good idea

STRAWPOLL: Proposal is to continue with the shortnames proposed in the Github issue #4, with the addition to rename SPARQL-overview to SPARQL-concepts

<Souri_> I had to get out of VPN and now it works

<pchampin> https://github.com/w3c/rdf-star-wg/issues/4

<gkellogg> +1

<ora> +1

<enrico> +1

<gtw> +1

<AndyS> +1

<pchampin> +1

<Souri_> +1

+1

AZ: +0.5 (not convinced about the SPARQL-overview to SPARQL-concepts)

TallTed: +1 strawpoll. Will probably be +1 to proposal based on final results of renaming discussion in the issue.

Doerthe: +1

DominitkT: +1

<Souri_> I have two connections: Souri and Souri_. How do I get rid of one?

ora: This looks in favour, by large

tallted: there is a missmatch between "SPARQL 1.1 Graph Store HTTP Protocol" and the IRI of the document which is named "http-rdf-update" right now
… this could lead to confusion in the new spec. I propose to rename it to Graph Store Protocol in the IRI as well

AndyS: this was renamed during the creation of the spec
… (Graph Store Protocol)

<gkellogg> Let's reduce the dependency on HTTP in names and specs, to the degree we can.

<Souri> For uniformity, my vote will go towards RDF-RDFXML.

ora: I'm in favour of graph store, on the fence on XML. I can be either way.

<pchampin> ghurlbot, names off

<ghurlbot> pchampin, OK.

ora: TallTed will do an updated proposal and we discuss it on the mailing list

ora: Peter Patell Schneider wrote a mail on the list

Peter Patel-Schneider's message on the mailing list

enrico: summarizes what is stated in the mail

<pchampin> https://lists.w3.org/Archives/Public/public-rdf-star-wg/2022Dec/0056.html

tallted: I do not see any mention on things that were already discussed. We spent a year on things that are discussed here, I don't want to get back to that.

Souri: I agree with what Ted is saying. No one wants to completely remodel their data once something changes.

enrico: I disagree with Souri. RDF is not just a data model, it is a knowledge representation language, grounded in model theory. If people have used it in bad ways, we should not encourage them to keep doing it

ora: my conclusion from this mail is we don't need to go into the dicussion of modal logic.