– DRAFT –
RDF-Star WG biweekly meeting

Approval of minutes from the last two meetings: 1 , 2

<AZ> In the minutes of meeting 13th June, we can add myself in the "regrets" section

<pfps> both minutes look fine

ora: Minutes again, anybody has a concern?

<ora> PROPOSAL: Accept the last two meetings' minutes?

<gkellogg> +1

<ora> +1

+1

<AndyS> +1

<TallTed> +1

<eBremer> +1

<niklasl> +1

<gtw> +1

<ktk> +1

<pfps> +1

<AZ> +1

<doerthe> +1

<tl> +1

<Dominik_T> +0 (absent)

RESOLUTION: Accept the last two meetings' minutes

<Kurt> +1

ora: side note, I am happy that we see 16 people, it's excellent

<Kurt> https://github.com/w3c/rdf-star-wg/wiki/Proposal:-Named-Node-Expressions

No meeting 4th of July

Kurt: I have a proposal ^ and ask to have some time to discuss it

ora: Everybody can read the proposal

ora: Next week is 4th of July, public holiday in US. I do not intend to work and I guess other people too. The logical thing would be to cancel the meeting. Thoughts?

<pfps> some of us in North America do not celebrate the fourht

ora: It means we shift the schedule

<tl> btw, will we have a pause for western hemisphere summer vacations?

ora: I think we can cancel the meeting

<pfps> I plan on being available throughout August

ora: I am availlable most of August

ktk: We have not done formal breaks

ora: should we do a poll

ora: let's do a poll

Proposal for next week's discussion

AndyS: at the semantic task force we had a good discussion. I can make a summary

ACTION: pchampin Adjust the focused meeting agenda in regards to the skipped meeting on July 4th.

<gb> Cannot create action. Validation failed. Maybe pchampin Adjust the focused meeting agenda in regards is not a valid user for w3c/rdf-star-wg?

AndyS: The observation is that in the general case if we have to deal with opaque URLs, if we consider that RDF is a web data format with the rules of URIs are used, the issues around URIs are less important: eithers the URIs are fresh to you or you adopt them from third pary and you adopt what they mean

<niklasl> +1 a good summary

AndyS: It would be good if we could talk about the usecases

<gkellogg> +1

<AndyS> May'24 semantics: https://github.com/w3c/rdf-star-wg/wiki/RDF-star-%22minimal-baseline%22

<niklasl> +1 also on use cases

ora: I think we are implicity in agreement

gkellogg: I think the conclusion we were comming to is that there is no strong support to do anything but transparent terms

<ktk> PROPOSAL: Next focus meeting topic: Respect the conventions on the web for use of IRIs

<ora> +1

<niklasl> +1

<ktk> +1

<gkellogg> +1

+1

<tl> +1

<TallTed> +1

<doerthe> +1

<eBremer> +1

<AndyS> +1

<gtw> +0

<AZ> +0

RESOLUTION: Next focus meeting topic: Respect the conventions on the web for use of IRIs

ora: Let's do one more discussion on how should we discribe the disconnect between asserted and unasserted triples

tl: I can try to give input in a not too long email

<fsasaki> +1 to a summary, would help a lot

gkellogg: We are planning to address Kurt's proposal as well

ktk: we might consider doing a longer meeting if we have a reason for it

ora: there is a chance I won't make it

<tl> unasserted assertions are ambiguous and would profit from a redesign

Named Node Expressions Proposal

ora: we can treat it as a "background" topic

tl: I noticed the usecases do not reflect the state of our discussion with stats and occurences

<tl> okay with discussing this in tomorrows semantics/usecase TF meeting

Review of pull requests, available at 3

<Kurt> Could someone please post the invite for the Semantics meeting to the IRC?

gkellogg: we agreed to merge the rdf:JSON proposal but Peters has more comments

gkellog: I would prefer to merge the PR now and work on subsequent PRs

pfps: The PR is not satisfactory. The value space and value mapping are unsatifying

pfps: I can write a mapping that satifies a document but produces '"b"' from the input '"a"'

pfps: The mapping is not trivial. It gots weird things related with UTF-16

gkellogg: I asked for help because I did not feel I am able to move this PR forward. It is the one PR with the most comments. I don't see how we move forward from here

TallTed: perhaps it can be step resolving this if Peters can open an issue for each problem he sees in the PR and be very explicit about each issue

TallTed: I think we have a hope to handle it 1 by 1. Even if we don't manage to solve them all we can at least track the problems

TallTed: If we don't do that I am afraid we will move in circles

pfps: That's not a way to run a standards group at all

pfps: I responded today with a fairly long writing about where the problems are in the current PR

pfps: The current lexical mapping does not say what is the mapping for string

ora: Do we have to find someone outside the group to help us with this?

ora: We could find somebody

ora: This starting to become a quite large issue. It's hard to me to see if this is essential for this group to solve

<Kurt> Can someone post the document link about rdf:json? I've been working with the JSON document object models for a couple of decades.

<AndyS> w3c/rdf-concepts#66

<gb> Pull Request 66 Updates rdf:JSON value space. (by gkellogg) [spec:substantive]

<Kurt> Thank you.

gkellogg: There is room for improving mappings. If people have objections on text, an efficient way is to suggest an alternate wording

gkellogg: The JSON-LD working group had a look at it yesterday and agreed with the current state

ktk: I cannot comment on the content but we made an agreement two weeks ago to merge it

ktk: Sadly Peter you got droped when we discussed if you would be fine with it

ktk: I think we should merge this PR in its current form

pfps: I am willing to drop my request for changes. I can try to approve

<gkellogg> Approve using "Review Changes" in https://github.com/w3c/rdf-concepts/pull/66/files

<gb> Pull Request 66 Updates rdf:JSON value space. (by gkellogg) [spec:substantive]

pfps: I am fine to take the last comment and turn it into an issue but there current version is not acceptable for an RDF datatype

TallTed: I hear you that this is not acceptable at as final result but I believe that this PR makes the document in a better state. It's incremental progress

gkellogg: The JSON-LD added the datatype to the rdf: namespace after a comment on the mailing list

ora: where are we with this entailment algorithm

Issue Triage, available at 4

pfps: I have finally found how to approve the rdf:JSON PR

gkellogg: About w3c/rdf-concepts#86 (datatype canonical mapping). We decided we do not need to provide a canonicalization for rdf datatypes

<gb> Issue 86 canonical mappings for datatypes (by pfps) [needs discussion] [spec:enhancement]

<pfps> I'll put together a PR for the canonical mapping issue

<pfps> Feel free to give me an action on this

ora: What RDF canonicalization does on datatype

<Kurt> https://www.w3.org/community/ixml/2021/03/19/welcome-to-ixml/

AndyS: It does not canonicalize

Kurt: I know Steven Pemberton have worked on the ixml spec to convert datatype to specific output. It may be possible that conversation with Steven could be productive