Meeting minutes
Approval of minutes from the last two meetings: 1 , 2
<ktk> Comments by az: https://
ora: any comments on the minutes?
ktk: we have a comment by AZ on the mailing list, link above
ktk: maybe the people mentioned in this email can tell pchampin which changes to make
TallTed: the minutes are not a meticulous transcription; "gibberish" and "poetry" happen.
ora: we might still address points if they are very unclear
<AZ> I agree that nothing I mentioned is critical
pchampin: do we want to set a deadline for addressing this issues?
TallTed: the bare "+1"s are the scribe scribing
gkellogg: the phantom issue is still there
ora: let's give people until next Monday to suggest fixes; can we do a conditional approval in the meantime?
pchampin: I think we can
<TallTed> pchampin -- if you can change `does this there is mean the specs are still still` to `does this mean the specs are still`, that will fix the poetry
<ktk> PROPOSAL: Accept last week's and two week's ago minutes, given additional feedback by next Monday (22.4.24). https://
<AZ> If no feedback is given, I'd be happy to approve the minutes as they are
<ora> +1
<ktk> +1
<pchampin> +1
<olaf> +1
<niklasl> +1
<gkellogg> +1
<AZ> +1
<tl> +1
<Dominik_T> +0.5 (I was not present on April 04, 2024)
<Souri> +1
<TallTed> +1
<TallTed> s| PROPOSAL: Accept last week's and two week's ago minutes, given additional feedback by next Monday (22.4.24). https://
RESOLUTION: Accept last week's and two week's ago minutes, given additional feedback by next Monday (22.4.24). https://
Proposal for next week's discussion 3
<tl> Page Not Found
<TallTed> s|PROPOSAL: Accept last week's and two week's ago minutes, given additional feedback by next Monday (22.4.24). https://
ktk: the discussions were super interesting; I have no idea how we are going to finish them.
… I'm a bit confused who is in which camp.
… Some positions are very clear, others are not so clear.
… Maybe we need to make a straw poll.
<tl> that link - https://
ora: for the sake of trying to solve this, I'm hoping that we only have two camps.
… I think we want to reach consensus, avoid an impasse.
… Hopefully we need to find consensus.
pchampin: like the other time, I think we should come to the next meeting with thoughts on "what I can live with" rather than "how can I convince the other camp".
… and set ourselves the goal to end the next meeting with a decision, which makes everyone equally (un)happy.
ora: I like this proposal.
gkellogg: apologies, I will not be here the next two weeks.
… I think there are really 3 positions, not 2. If we consider the possibly entailed triple terms, this leads to several triple terms for a single reifier.
… We need to consider ways to fix our decision if we need to.
ora: what kind of entailment are you thinking of? sameAs?
gkellogg: there is also cases of equality with numeric literals.
Souri: maybe we need to think about what we can live with, but also what we absolutely can not have.
… It's good to have arguments for both.
tl: a Best Practices document would be a good place to explain what can be done.
… I understand that niklasl was working on such a document. I would like to know the status.
niklasl: it's hard to say. It is not ready yet, but maybe it can already be shared.
AndyS: Can we have material for the next meeting several hours before the meeting, please.
… It is hard to catch up with material just before the meeting.
ora: agree. What about setting the limit to the previous day?
Review of open actions, available at 4
gkellogg: I have updated all the specs to be "updatable".
… They have been merged, but not published, because of a bug with the github action.
<gkellogg> close w3c/rdf-star-wg#113
<gb> Closed issue #113
pchampin: there was indeed a bug in the GH action that we are using; it was fixed in the meantime, so rerunning the actions should fix the problem.
Review of pull requests, available at 5
ora: I guess there is nothing we can do about these either.
… How can we solve this rdf:JSON issue? pfps is not here.
… The term is here. The question is a matter of taste on how we define the value space.
… I can't speak for pfps.
… I think we need a focused meeting to discuss this issue.
ora: from your experience with JSON-LD, how would you solve it?
gkellogg: in JSON-LD we relied on JCS canonicalization, even though JCS was not an RFC.
… I interpret pfps' feedback as considering that JCS is inadequate -- it can be lossy in some cases.
… If we were to create a new canonicalization scheme, using XML-Schema canonicalization, this would solve this problem.
… But that would be yet another JSON C14N algorithm...
… But the difference between the two would not have any practical impact in my opinion.
<pchampin> +1
ora: we should soon have a discussion about this, so that we can put it behind us.
Issue Triage, available at 6
<gkellogg> w3c/
<gb> Issue 114 How will RDF 1.2 affect RDF Canonicalization? (by niklasl) [needs discussion]
gkellogg: in my opinion, we need to define a transformation from graph using triple-terms to graphs without triple-terms,
… ideally a round-trip-able transformation,
<niklasl> +1
gkellogg: and define that the canonicalization of RDF 1.2 as the canonicalization of that transformation.
… There was a lot of work put in the C14N to prove its good properties from a cryptographical point of view.
… Any change to include the triple-terms would change it radically.
ora: I understand we don't want to open that long discussion again.
… Do we need to dedicate one of the focused meeting on this?
gkellogg: I would like to hear someone with a different view. We need to discuss the technical means to define this transformation?
pchampin: I agree with gkellogg's proposal, although I would frame it clearly as a workaround, pending a new version of RDFC.
AndyS: I don't have a different opinion either. About round-trip, I don't think we need an RDF 1.1 -> RDF 1.2 -> RDF 1.1 roundtrip.
niklasl: I agree with the other remarks.
… [something about triple-term identifiers]
ora: any other issues that someone wants to discuss?
<gkellogg> I was thinking of round-tripping RDF 1.2 Full to RDF 1.2 Classic.
ora: I suggest every one takes a look at the list every so often.
Any Other Business (AOB), time permitting
ora: who is going to the Knowledge Graph conference? https://
… I am. It's in NYC in 3 weeks.
niklasl: I can't go, and probably not even watch it remotely.
… Will they record your presentation?
ora: I think they record all the talks. I'll do a dry-run internally, it might be recorded as well (in case the KGC does not).
… There will be a lot of interesting stuff, inc. a master class in SHACL.
… Anybody going to ESWC?
olaf: I am.
<ktk> zazuko/
ktk: Knowledge Graph Forum in Basel, ora will be here.
<TallTed> zazuko/
ktk: Anybody wants to come, let me know.
… It is back-to-back with Semantics @ Roche. https://
ora: I'm going to Finland after that, I was invited to a semantic meeting.
… They asked me to talk about RDF-star.
ora: let's adjourn. Please remind to send any input to the next meeting at least 24h before.