– DRAFT –
RDF-star WG

<pchampin> https://www.w3.org/2001/12/zakim-irc-bot.html

<pchampin> https://www.w3.org/2002/03/RRSAgent

<pchampin> https://w3c.github.io/scribe2/scribedoc.html

<Enrico> p+

<ora> PROPOSED: Approve last week's minutes https://www.w3.org/2022/12/08-rdf-star-minutes.html

<pchampin> +1

+1

<ora> +1

<ktk> +1

<Dominik_T> +1

<gtw> +1

<AZ> +1

<TallTed> +1

<Enrico> +1

<AZ> But we need to approve the minutes of all the past meetings

<AndyS> +1

RESOLUTION: Approve last week's minutes https://www.w3.org/2022/12/08-rdf-star-minutes.html

<pchampin> previous minutes: https://www.w3.org/2022/12/01-rdf-star-minutes.html

<pchampin> ... https://www.w3.org/2022/11/17-rdf-star-minutes.html

<pchampin> ... https://www.w3.org/2022/11/10-rdf-star-minutes.html

<ora> PROPOSED: Approve earlier meetings' minutes

<ktk> +1

<ora> +1

<Dominik_T> +1

<pchampin> +1

<Enrico> +1

+1

<gtw> +1

<AZ> +1

<TallTed> +1

RESOLUTION: Approve earlier meetings' minutes

ora: agenda items ...
… discussion on the public m.list
… home work assignments
… naming of the documents
… Let's begin with the naming

Naming of the specifications

pchampin: sent email to the mailing list
… current naming of RDF RECs is a mess

<pchampin> https://lists.w3.org/Archives/Public/public-rdf-star-wg/2022Dec/0024.html

pchampin: b/c long history and high number of docs
… created mapping
… propose to prefix all specs with 'rdf12-'
… and in favor of having 'rdf-*' links point to the latest versions of the specs
… unless there is a reason for multiple versions to cooexist, which is not the case at the moment

ora: Thanks for creating the diagram!

<pchampin> https://www.w3.org/TR/html4/

TallTed: nowadays superceeded docs need to contain a note saying that it is superceeded

pchampin: yes, HTML4 spec is an example

<Zakim> gtw, you wanted to ask if that rdf12- prefix is suggested even for sparql specs

pchampin: and we should do that for consistency

gtw: was the proposal about 'rdf12-' for all specs (incl. SPARQL-related ones)?

pchampin: no, not necessarily for the SPARQL-related ones
… but the question may be raised for Turtle and other serialization formats

<gtw> maybe we could have redirects /turtle -> /rdf12-turtle

AndyS: we can have redirects, as suggested by gtw in the chat

ora: My question was what happens if we choose this naming convention now, how much trouble would we have if we realize later that it wasn't the best decision

pchampin: we don't need a strong decision now
… but once we publish the first public working draft

ora: we shouldn't take this decision lightly

<TallTed> "SPARQL" being "SPARQL Protocol and RDF Query Language" I suggest we *do* use the rdf12- prefix on it, and ensure that all specs we touch are synchronized (unlike SPARQL 1.1 being based on RDF 1.0, while RDF 1.1 has no query language yet)

pchampin: two questions actually 1/ how do we name our new versions 2/ ... (?)

TallTed: all specs should be synchronized in terms of naming, including the SPARQL-related ones

AndyS: SPARQL 1.1 was designed for both RDF 1.0 and RDF 1.1 -- i.e., if you feed RDF 1.0 into it, you get RDF 1.0 out; if you feed RDF 1.1 to it, you get RDF 1.1

TallTed: that's not clear from the spec

AndyS: there was no RDF 1.1 at the time

ora: I like TallTed's suggestion to prefix everything with 'rdf12-'

<AndyS> SPARQL 1.0 :: https://www.w3.org/TR/rdf-sparql-query/

ora: any objections?

AndyS: Would we put redirects for SPARQL-related naming convention?

<TallTed> I would redirect from https://www.w3.org/TR/rdf-sparql-query/ to https://www.w3.org/TR/rdf12-sparql-query/ and move the current https://www.w3.org/TR/rdf-sparql-query/ to https://www.w3.org/TR/rdf10-sparql10-query/ (or sparql11)

AndyS: expectation for SPARQL might be that the next one would be sparql12-query

ora: no conflict with new naming convensions if we add redirects

pchampin: official name is REC-...
… the short names are already aliases only

ora: so, short names could be redirected already now without causing problems for expectations for the old versions

<pchampin> https://www.w3.org/standards/history/rdf-schema

<gtw> I would be more comfortable if we had an explicit list of the proposed names, not just a wildcard pattern.

pchampin: history of short names
… convenient for browsing the old versions

<ora> PROPOSED: Adopt spec naming scheme which prefixes all our docs with "rdf12-" + add redirects from existing short names to the new docs

<pchampin> +1

<TallTed> +1

<ora> +1

<Dominik_T> +1

ora: proposal to adopt naming scheme to prefix all our docs with 'rdf12-' and add redirects for the short names

gtw: tension between the SPARQL-related names and the RDF-related names
… because there was not always a 1:1 mapping between the version
… e.g., there is a SPARQL 1.2 CG at the moment

ora: we are taking over docs
… whatever the name of that doc, we prefix it

<TallTed> the SPARQL 1.2 CG (and its repo) can be renamed/moved

gtw: but some of these docs already have version numbers

pchampin: shouldn't much of a problem if we make these standards open to adding new features
… "living standards"
… i.e., same REC with the same version but with more features added
… if we go in this direction, not much of a problem for the SPARQL 1.2 CG
… because they wouldn't need to start a new version

gtw: so, two version numbers in the name of the updated SPARQL spec?

pchampin: no, that was not how I understood it

ora: think of it like "forks"
… but maybe we need to "sharpen" the definition that I was proposing
… like the idea that the only version number in the name is the 1.2 for RDF

<ora> ack +

pchampin: system might not even support/permit two version numbers in a name

ktk: same concerns at gtw
… last time we said we like the overview/entry doc that SPARQL has
… and we would do something similar for the RDF-related docs

<pchampin> in a way, the RDF primer plays a similar role to the SPARQL overview doc

ktk: if we have that, this may avoid such issues about the version numbers in the names

<pchampin> https://www.w3.org/TR/rdf11-primer/

<pchampin> https://www.w3.org/TR/sparql11-overview/

ktk: prefer to have the version number in the names of all docs

ora: in this case, it may be useful to have that overview doc first
… including a list of all the docs that we will have
… let's have such a list first
… then, we can discuss it on the mailing list, including discussing possible names for them

ACTION: pchampin to write a full proposal of all document names

<ghurlbot> Created action 55 write a full proposal of all document names (on ) due 22 Dec 2022

pchampin: I can make a proposal of such a list

<AndyS> Please include the *-12 style.

ora: from a W3C perspective, better to make this decision now rather than getting into trouble later

AndyS: to pchampin, please also include the options with the '-12' in the end of the names

What are big missing things from the CG spec?

ora: we need some vocab so that we can use RDF-star in a schema
… I would like to be able to define an RDFS schema so that I can use RDF-star

<pchampin> and SHACL shapes

ora: additionally, we need to make sure that we understand the ramifications of the formal semantics

ktk: Turtle is a subset of N3

<AndyS> +1 to SHACL (though it's not mentioned in charter)

ktk: with the current Turtle-star syntax we might break that?
… was/is that a concern?

pchampin: there is an implementation of N3 that includes RDF-star

ora: that topic should go on the list

pchampin: there is a CG for N3
… not same level of maturity
… we should liason with that group
… Doerthe is in that group

ora: yes, we can do that
… the point now, however, is a comprehensive list of things
… Enrico had some thoughts?

<pchampin> coming back to the vocabulary, there is a first attempt at https://www.w3.org/2021/12/rdf-star.html#rdf-star-vocabulary

Enrico: not clear to me if the motivation is focused completely on the reification use case
… example "I believe that John is dead" versus "I believe that the date of ... mariage ..."
… I believe that RDF-star is about reification
… not about believe use case ("I believe John is dead") because "John is dead" is not a resource
… the old examples of RDF* should not be the object of the discussion now
… there should be complete transparency in RDF-star
… not opacity
… I suspect that the reason for having opacity is about the confusion of ... (?)
… it should be the job of N3 to be about unasserted statements
… / believes

ora: the mailing list would be a could place to put these thoughts

ora: earlier reference to modal logical

Enrico: reification is something else

ora: Enrico, please compose an email to the mailing list to make clear what your thoughts are so that we can discuss them
… discussions about asserted and/or quoted triples are needed
… the home work stands, let's use the mailing list
… so that we can quickly get an understanding of the full scope of the work ahead
… we have a meeting next week

AndyS: since not everyone can come to next week's meeting, would that be a non-decision-making meeting?

ora: yes

<Souri> quit

<Enrico> quit

<TallTed> previous meeting -- https://www.w3.org/2022/12/08-rdf-star-minutes.html

<TallTed> next meeting -- https://www.w3.org/2022/12/22-rdf-star-minutes.html