W3C

RDF-star WG biweekly meeting

13 June 2024

Attendees

Present
doerthe, eBrermer, gkellogg, gtw, ktk, niklasl, ora, pchampin, pfps, rubensworks, Souri, Souri6, TallTed, tl, Tpt
Regrets
andys, AZ, enrico, fsasaki, olaf
Chair
ora
Scribe
rubensworks

Meeting minutes

<gb> MERGED Pull Request 57 Charter extension 2024-03 (by pchampin)

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

ora: Any concerns about the last minutes?

<ora> PROPOSAL: Accept 2024-05-23 minutes

<ora> +1

<ktk> +1

<TallTed> +1

<eBrermer> +1

<niklasl> +1

<pfps> +1

+1

<gtw> +1

<gkellogg> +1

<tl> +1

<pchampin> +1

<doerthe> +1

<Souri> +1

RESOLUTION: Accept 2024-05-23 minutes

<ora> PROPOSAL: Accept 2024-06-06 minutes

<ora> 0

<ktk> +1

<gtw> +1

<gkellogg> +1

<niklasl> +1

+1

<tl> +1

<pchampin> +1

<eBrermer> +1

<TallTed> +1

<doerthe> +1

RESOLUTION: Accept 2024-06-06 minutes

Re-chartering 3

<gb> MERGED Pull Request 57 Charter extension 2024-03 (by pchampin)

pchampin: A new PR was pushed about possible change of name. Changing name mid-charter is unusual, so I rephrased to to say that we may change the name in the next charter.
… Other than that, it should be ready to go.

ora: What do we need to do next?

pchampin: We need to go to the strategy team and to say what we want to submit. Already talked to them, so there should be no pushback. This will be a regular AC vote.

ktk: Is a formal resolution needed in this group?

pchampin: Let's make a resolution to be on the safe side.

<ora> PROPOSAL: Authorize pchampin to take the new charter proposal to the W3C strategy team

<ora> +1

+1

<gkellogg> +1

<eBrermer> +1

<pchampin> +1

<Souri> +1

<gtw> +1

<TallTed> +1

<ktk> +1

<niklasl> +1

<tl> +1

RESOLUTION: Authorize pchampin to take the new charter proposal to the W3C strategy team

<Tpt> +1

<pfps> +1

<Zakim> tl, you wanted to ask for Ora's trip report from Basel, especially the discussion with Hannah Bast (developer of the Qlever triple store) that Adrian mentioned last week

tl: I'm interested in the difficulties on RDF-star in Qlever.

ora: I didn't get the feeling they thought about it much.
… They don't support SPARQL 1.1 fully yet either.

tl: Maybe it's not important here.

ora: I explained to her what we are doing in the group atm.
… They are working through a bunch of stuff still in their SPARQL impl.
… I will email her.
… I was impressed with the performance of that system.
… They are orders of magnitude faster.
… There are no special tricks in the implementation. Just good understanding of the algorithms and clean code.

Proposal for next week's discussion

ora: enrico is working on a new proposal which simplifies things.
… I'm sure he will send it before next week's meeting.
… Next week's topic should probably be about that.

pchampin: Is this a new proposal next to the baseline?

<pchampin> https://www.w3.org/2002/09/wbs/139681/2024-07-baseline/results

pchampin: The results of the poll can be found at this link.
… This was to ask about the support for the baseline proposal.

ora: We should give him the opportunity to explain his proposal.

pfps: I thought we were at the point of voting, not to look at new things.

ora: I don't disagree.

pfps: We did vote.

TallTed: This is not a baseline, should be something we should grow from.

ora: What do you propose?

TallTed: Given that this is a strawpoll, this is not a binding decision, that decisions be clearly communicated.
… The strawpoll does not accomplish the goal.

ktk: I disagree.
… We should set on something that is the foundation. Further proposals should be framed on that proposal.

ora: Could we resolve something next week?

ktk: I don't know.

tl: I don't understand the need for a strawpoll at all.

<TallTed> more metamodelling crap and

<TallTed> more axiomatic triples crap

TallTed: The above is not part of a baseline.

<TallTed> all the RDF 1.1 metamodelling crap and

<TallTed> all the RDF 1.1 axiomatic triples crap

TallTed: Major changes must be made to this draft.
… I don't see how it could be made in a shape that we can vote on next week.
… Previous proposals were too open-ended, leading to incompat, just like this proposal.
… Feels like we're starting over.

niklasl: There is frustration that we're not progressing. I want to talk about the reqs and use cases.

<niklasl> https://lists.w3.org/Archives/Public/public-rdf-star/2024Jun/0000.html

<pfps> My bad - I should realize the difference between the two mailing lists.

niklasl: I abstained from voting, because I agree that we need something clearer to vote on.
… Opacity came back, and no one really wanted that.
… I tried to show the way I see where we're at in my reply to the mail.
… I agree we need to talk about something concrete.

ora: The word "baseline" is probably not the right one.

gtw: I agree with that. We're getting too hung up on the word "baseline".
… The instructions were clear on the strawpoll.
… Previous proposals were hard to compare because there was nothing clear to compare to.

<gkellogg> +1 to gtw

gtw: We need something to compare to as we move forward.

ora: This ability to compare is very important.
… This is still difficult for me.
… And how to compare this to the broader group of stakeholders.

<Souri> "benchmark" (instead of "baseline") for comparison whenever something new is proposed

ora: I propose next week we discuss the strawpoll results, to see if we agree on this "baseline".

tl: I want to understand what the minimal set is.
… E.g. RDF* (RDR) + occurrences.

ktk: What is the alternative if you disagree with the baseline?

pfps: My strawpoll vote had a proposal for another baseline. It's a subset of the page that we were voting on without the crap.
… We had use cases, but didn't spend time discussing them.
… If we evolve things in RDF 1.2, we should have approved use cases for them.

TallTed: We should have approved requirements that come out of the approved use cases.
… So we can compare and test which proposals meet the requirements.

<pchampin> +1

TallTed: I would like us to get to this.

niklasl: I agree. I want to help with the use cases.

<ora> PROPOSAL: Next week: validating the baseline against use cases

<ora> +1

<gkellogg> +1

<gtw> +1

<tl> +1

<ktk> +1

<eBrermer> +1

+1

<pchampin> +1

<niklasl> +1

<doerthe> +1

<Tpt> +1

<TallTed> +1

RESOLUTION: Next week: validating the baseline against use cases

ktk: I will add the link to the agenda for next week.

TallTed: It may be time to move that stuff from the wiki to managed pages.
… Treating it like a spec document, with PRs.

<ora> +1 to TallTed's idea

<pchampin> +1

<niklasl> +1

<ktk> +1

<eBrermer> +1

<Souri> +1

<gkellogg> +1 should be a WG Note in ReSpec format.

<doerthe> +1

gkellogg: You can make respec documents about markdown documents.

<TallTed> +1 WG Note in ReSpec

Review of open actions, available at 4

pchampin: If we have a new note, we should re-activate the workflows.

Review of pull requests, available at 5

gkellogg: There are some trivial PRs on Turtle and TriG.
… The big one is rdf:JSON.
… A note was added about infinite numbers.

gkellogg: It's a 10-11 month PR atm, so we need to move on.

pfps: It looks like rdf:JSON may actually be an RDF datatype now. At the expense of being different from other JSON.

gkellogg: Similar to xml and html literals, those groups saw the need to represent it in RDF. So did the JSON-LD group for JSON.
… We removed canonicalization from it.

pfps: The JSON-LD people should look at it.

gkellogg: The expectation from the JSON-LD group would be that the RDF WG would include it.

<TallTed> conveniently, gkellogg is a JSON-LD people. As am I.

<TallTed> might be worth a "call for objections from JSON-LD CG/WG"

pfps: Having this as a PR is no different than a draft. They are constantly changing.
… The JSON-LD WG should approve it before it gets in.
… We have a fair cross-representation between these 2 groups.

ktk: Can we live with merging, and asking for feedback from the JSON-LD WG?

niklasl: I think it would be good to merge now.

<pchampin> @pfps, by merging it, we generate a snapshot of the WD, so it is still more stable than a PR

ora: I like the idea to merge now, and solicit feedback. And if needed, create another PR.

ora: Anyone disagree?

<TallTed> we can ask JSON-LD people to make substantive change requests -- i.e., PRs and/or issues -- making clear their objection

pchampin: Peter needs to change his review before it can be merged.

<ktk> pfps -- the PR is blocked by you, you would have to unblock

<ktk> pfps -- we didn't hear you anymore on zoom

Summary of resolutions

  1. Accept 2024-05-23 minutes
  2. Accept 2024-06-06 minutes
  3. Authorize pchampin to take the new charter proposal to the W3C strategy team
  4. Next week: validating the baseline against use cases
Minutes manually created (not a transcript), formatted by scribe.perl version 221 (Fri Jul 21 14:01:30 2023 UTC).

Diagnostics

Succeeded: s/p+//

Succeeded: s/[]/enrico/

Succeeded: s/RDF-star + occurrences./RDF* (RDR) + occurrences./

Succeeded: s/pfps: the/pfps -- the/

Succeeded: s/pfps: we/pfps -- we/

No scribenick or scribe found. Guessed: rubensworks

All speakers: gkellogg, gtw, ktk, niklasl, ora, pchampin, pfps, TallTed, tl

Active on IRC: doerthe, eBrermer, gkellogg, gtw, ktk, niklasl, ora, pchampin, pfps, rubensworks, Souri, Souri6, TallTed, tl, Tpt