W3C

– DRAFT –
RDF & SPARQL WG

23 October 2025

Attendees

Present
AndyS, AZ, doerthe, Dominik_T, fsasaki, gtw, ktk, lisp, niklasl, ora, pfps, Souri, TallTed, tl
Regrets
olaf, pchampin
Chair
ora
Scribe
doerthe, ktk

Meeting minutes

Approval of last week’s minutes: 1

<ora> PROPOSAL: Approve last week's minutes?

<ora> +1

<AndyS> +1

<gtw> +1

<TallTed> +1

<doerthe> +1

<ktk> +0 (absent)

<niklasl> +1

<lisp> +1

<tl> +0 (absent)

RESOLUTION: Approve last week's minutes?

Dependency graph of our specs 2

ora: we discussed this in the chair meeting, pa made a dependency diagram of our specs

AndyS: this raises the questions: in which way should be read the spec?

ora: There is one arrow which produced circles

<niklasl> w3c/rdf-concepts#250

<gb> Issue 250 Cyclic dependency between Concepts and Semantics (by pchampin)

AndyS: there is already a PR that fixes it
w3c/rdf-concepts#250

<AndyS> w3c/rdf-concepts#251

<gb> MERGED Pull Request 251 remove normative dependency to Semantics (by pchampin)

ora: is there an actual PR?

AndyS: yes, w3c/rdf-concepts#251

<lisp> the same gpt conversation that generated the detail graphs also include its reading of the documents : https://chatgpt.com/share/68f8c192-b3f8-8001-aef6-dcf471177b69

ora: so the normative dependency from concepts to semantics is removed, that serves the purpose

ktk: I think AndyS is right, that we should give some guidance to help people how to read the spec

<niklasl> Tangential, this is a good approach to write documentation (for different purposes): https://diataxis.fr

ora: Maybe the dependency graph can serve as a landing page. I think it is best to start with the primer.

ora: I think we are OK with the normative dependencies now

TallTed: There were 3 issues flagged with the dependency graph and only one of these is solved.
… maybe it is only the image?

ora: so the arrow from semantics to concepts is removed? But then there is still a cycle, right?
… or did I misunderstand that?

<niklasl> See https://www.w3.org/TR/rdf12-semantics/#normative-references (RDF12-TURTLE is there)

ora: did the arrow from semantics to concepts was removed?

<niklasl> Nothing to semantics now in https://w3c.github.io/rdf-concepts/spec/#normative-references (it's gone in the latests editors draft)

AndyS: I think it is not possible to break the cycle. The connection from concepts to semantics is rather weak, the link is just that the semantics is defined in semantics

ktk: so the solution is that there is not connection from concepts, right?

niklasl: the other dependency is removed, we should be fine

ktk: it is strange that n-triples depends on turtle, but I think that was for historical reasons

ora: I think that is the case and it can be explained
… is it right that TriG does not depend on turtle?

<niklasl> This is the script PA made to do the analysis: https://github.com/pchampin/w3c-dep-graph/blob/main/gendot.py

AndyS: yes, there can be still non normative dependencies

ora: I think there are no technical problems, no cycles, so we can go on.

Review of open actions, available at 3

ora: my task is done, but there is still an echidna issue

ktk: line item 1 and 2 can be closed

ora: where are we with the owlification thing?

Horizontal Reviews, last feedback on questionaire 4 5

niklasl: I still need to get some feedback from pa, but everyone can already review, you find it in my repo

ora: can we close the dependency graph issue

ktk: I would like to wait for pa

<AndyS> w3c/rdf-concepts#252

<gb> Pull Request 252 complete description of the main changes (by pchampin)

AndyS: I think that issue #183 can be merged, but we need to wait for pa

<gb> Action 183 add a summary of the main changes since RDF 1.1 in rdf-concepts (on pchampin) due 2025-10-23

Identifying issues to solve before CR 6

ora: we discussed in the chair's meeting that we will submit the questionnaire for all documents together and not separately (agendum 4)

ktk: we should decide which document tackle next, turtle?

<AndyS> w3c/rdf-concepts#255

<gb> Pull Request 255 RDF Reference IRI (by afs)

AndyS: I made the PR w3c/rdf-concepts#255
… which fixes issues #223 and #169

<gb> Issue 223 not found

<gb> CLOSED Issue 169 definition of reifiers is non-normative and seems vague (by rat10) [ms:CR]

AndyS: issues w3c/rdf-concepts#223 and w3c/rdf-concepts#169

<gb> Issue 223 RDF Reference IRI (by afs) [ms:CR]

<gb> Issue 169 Relative IRI Reference should bind to the irelative-ref production (by gkellogg) [ms:CR] [spec:editorial]

ktk: .. and the turtle issue can be addressed afterwards?

AndyS: yes

tl: AndyS, could you explain the relationship of the PR and the issue?

AndyS: It changes a production rule in the grammar

tl: but how does it relate to 169?

AndyS: are you looking at the correct 169?

tl: then it is clear, sorry

ktk: anything else to discuss?

ora: so turtle is what we tackle next, but also the other syntax specs
… does the fact that there is nothing on rdf/XML on the list that everything is fixed there?

<niklasl> +1 that's my recollection too

AndyS: I think things are fixed there

ora: does that mean that it is ready to go?

ora: we only need to contact jervan

<ktk> https://github.com/orgs/w3c/projects/20/views/11

doerthe: the old spec already contained a part where the interpolation lemma was in it.
… I will add it there

<ktk> https://github.com/orgs/w3c/projects/20/views/5

ltk: we could discuss/review https://github.com/orgs/w3c/projects/20/views/5

ora: what about the missalignment of prefix handling, where are we there?

AndyS: I don't remember

ora: there are a couple of issues about turtle, what do we do with these?

<ktk> w3c/rdf-turtle#109

<gb> Issue 109 Ambiguous wording in section 2.11.1 Annotation Syntax (by jaw111) [spec:bug] [spec:editorial]

<niklasl> It

<niklasl> It's labelled editorial, which seems correct (it's about possible clarification).

ktk: could we check whether this issue needs our attention? w3c/rdf-turtle#109

ora: I think we only need to add some clarifying text to fix it, do we think the text is needed?

AndyS: we cannot satisfy every single reader, so we have to be careful here

ora: I suggest to close the issue

TallTed: it might be that the person who posted it is not the only one having the problem

ora: I propose that I review and then report

TallTed: the problem comes from the annotation syntax and its text about reifiedTriple production, so I propose to rewrite the text, I will have a look whether I can come up with an improvement

ktk: most issues are on SPARQL, once the SPARQL spec is ready many issues here will be solved

<Dominik_T> to remember w3c/rdf-schema#65

<gb> Issue 65 update editor list (by domel) [spec:editorial]

<ktk> s/missunderstandings/misunderstandings/

Summary of resolutions

  1. Approve last week's minutes?
Minutes manually created (not a transcript), formatted by scribe.perl version 246 (Wed Oct 1 15:02:24 2025 UTC).

Diagnostics

Succeeded: s/1?//

Succeeded: s/183 can also be closed,/183 can be merged,/

Succeeded: s/aggendum/agendum/

Succeeded: s/annotation syntax and the text about it/annotation syntax and its text about reifiedTriple production/

Failed: s/missunderstandings/misunderstandings/

Succeeded: s/questionaire/questionnaire/

Succeeded: s/garph/graph/

Succeeded: s/the arrow from semantics to concepts removed/the arrow from semantics to concepts was removed/

Succeeded: s/were are we there/where are we there/

Succeeded: s/Its labelled editorial/It's labelled editorial/

Succeeded: s/wether/whether/

Maybe present: ltk

All speakers: AndyS, doerthe, ktk, ltk, niklasl, ora, TallTed, tl

Active on IRC: AndyS, AZ, doerthe, Dominik_T, fsasaki, gtw, ktk, lisp, niklasl, ora, pfps, Souri, TallTed, tl