Meeting minutes
<pfps> minutes look fine
Approval of last week's minutes: 1
ora: any comments on last week's minutes?
[crickets]
<ora> PROPOSAL: Accept last week's minutes
<AZ> +1
<ora> +1
<Dominik_T> +1
<ktk> +1
<pfps> +1
<gkellogg> +1
<rubensworks> +1
<TallTed> +1
<olaf> +1
<afs> +1
<enrico> +1
RESOLUTION: Accept last week's minutes
Review of open actions, available at 2
ora: nothing completed; pchampin is on vacation
Review of pull requests, available at 3
ora: we have a bunch of PRs to discuss. who wants to start?
gkellogg: concepts update and terminology...
gkellogg: I was referring to rdf-comcepts#32, but doing others first may make more sense
<ghurlbot> Issue 32 [not found]
ora: 2 editorial changes marked with need discussion...
<Zakim> pfps, you wanted to comment on LI 38
w3c/rdf-schema#8
pfps: it's just languishing. someone needs to work on it.
ora: any volunteers to fix it?
pfps: a number of us have proposed changes.
TallTed: Loooking at rdf-schema#8, I don't see any requested changes that have not been addressed
<ghurlbot> Pull Request 8 change range of rdf:predicate + small HTML fixes (domel) needs discussion, spec:editorial
TallTed: Other than requests to change prose. If you have a request, please make a specific request for a change.
TallTed: You have a proposed suggestion, but no specific wording suggested.
pfps: Can anyone suggest editorial changes to documents?
TallTed: Yes, anyone can.
afs: Commits can also be added to an existing PR.
<pfps> I'll put together a commit to resolve the issue, and change the issue title
gkellogg: you can make a PR against a PR (or rather, against the branch behind the PR)
… the suggestion mechanism is quite useful
… none of these are limited to editorial changes
TallTed: While a PR is a change request, be careful of loosing the chain of commits to the PR.
<pfps> I am again confused as to what I should do
ora: I'm suggesting that pfps take the lead on these changes
pfps: I don't know what I'm supposed to be doing, as GitHub actions
<pfps> OK, I'll push a commit on the branch of the PR?
<pfps> Or not?
<afs> I'm happy with commits on PRs I send in when the PR is otherwise quiet and the changes do not fit in a suggestion. e.g. the changes are in multiple places.
w3c/rdf-schema#9
ora: why is there needs-discussion on w3c/
gkellogg: pfps took issue with something there
pfps: I would like PR and Issue names and descriptions correspond to their content
gkellogg: you need to clear your change request, because that blocks it from being merged
<gkellogg> w3c/
<gkellogg> w3c/
w3c/rdf-concepts#32
gkellogg: we have not settled on what to call triples that don't themselves include quoted triples, vs those that do
… "simple" and "complex" were one suggestion, which I've started to apply
… we need to decide on nomenclature to use going forward; this was one attempt to do that
ora: any other comments?
afs: I'm not seeing the need for this extra terminology. a triple has 3 components, whether one of those may be a quoted triple or not.
… making this another concept seems unnecessary and potentially confusing
gkellogg: we do still have the collective "triple" including both "simple" and "complex"
afs: I don't think we need this in top-level terminology
ora: so we could have this elsewhere, not at top-level?
afs: yes
[back and forth between afs and gkellogg]
ora: so we have triples (all encompassing), complex triples (containing quoted triples as subj or obj), and simple triples (no quoted triples in play)
… could we have complex and non-complex triples, clarifying the disjoint relationship between these?
olaf: I can see both sides. I think the differentiating terminology will be useful, but I want nested and non-nested triples, instead of complex and simple
pfps: I wonder why we're discussing this in context of a directed PR, rather than an overarching issue
enrico: I think triples remain triples, and when used as terms (subj or obj), they are "triple-terms"
ora: how would you refer to the triples which use triple-terms?
enrico: RDF 1.1 triple, or simple/complex triple, maybe
gkellogg: to answer pfps, the discussion has arisen on a PR because the PR made the issue visible
… nested vs unnested might be reasonable alternative
<ora> "There are only two things that are hard in CS: naming, cache invalidation, and off-by-one errors."
gkellogg: retrospectively, we know that triples that include blank nodes do not need to be differentiated from triples that do not include blank nodes, but we don't yet have hindsight about nested/quoted/embedded/etc. triples
afs: what about changes in RDF 1.2 that are not about quoted triples? there's at least one such
… there seems to be a lot more editing needed
… I'm concerned about inventing terminology that will not be used except by confused people
… there hasn't been much discussion on the PR because it's labeled as a Use Case
enrico: we say that "ground triples" are triples without bnodes. maybe we say "flat triples" are those without embedded/quoted/nested
ora: clearly, naming is extremely important. at some future point, we will need a careful review to make sure that terms are used consistently.
… in history of RDF development, we used "pumpkin" as a stand-in for "nodes", which we hadn't figure out a name for yet
… we have two things: 1) what to do for this PR, if anything; 2) the broader issue of making all specs consistent internally and across the board
gkellogg: I think the controversial terms serve the purpose for now. afs has some ideas about doc organization which would help clarify things as we go forward
… specific comments in the PR will help move it along. terminology may be considered placeholders for now, with refinement to happen later.
ora: I think we've got a reasonable path forward for now
… time is short again, but perhaps we have time for one more item
<Zakim> pfps, you wanted to ask whether this PR will be merged in before FPWD
afs: are we adding material from CG report to these documents?
pfps: how does this relate to FPWD?
gkellogg: waiting until things are perfect will mean we never publish FPWD
… text direction, canonicalization, some other things want work
w3c/sparql-protocol#13
ora: what is w3c/
rubensworks: I created that
… it's just an informative addition to the SPARQL protocol spec, showing another SELECT query
<Zakim> gkellogg, you wanted to discuss disposition of other bits of the RDF-star CG report
brief AOB
afs: i18n issues are awaiting a consensus proposal, which I have floating around in my head, waiting to put forward
gkellogg: there are several elements of the CG report that need incorporation, some in rdf-new, some elsewhere
afs: question was about a principle decision, of whether we incorporate CG report into 1.2 specs now or later, after we resolve more basic updates than nested (or whatever)
gkellogg: much should/must wait until semantics TF has done its work
ora: we're at time
… propose adjournment
[no objections]
<ghurlbot> Pull Request 13 Add informative quoted triples example, Closes #12 (rubensworks) needs discussion
<TallTed> s| Pull Request 13 Add informative quoted triples example, Closes #12 (rubensworks) needs discussion||
<ghurlbot> Action 12 [closed] add ora's and greg's regret and chair name in last calls minutes (on pchampin) due 2 Feb 2023
<TallTed> s/s| https/s|https/
<TallTed> s| Action 12 [closed] add ora's and greg's regret and chair name in last calls minutes (on pchampin) due 2 Feb 2023||