14:49:04 <RRSAgent> RRSAgent has joined #rdf-star
14:49:04 <RRSAgent> logging to https://www.w3.org/2021/05/14-rdf-star-irc
14:49:07 <Zakim> RRSAgent, make logs Public
14:49:08 <Zakim> please title this meeting ("meeting: ..."), pchampin
14:49:17 <pchampin> meeting: RDF-star
14:49:20 <pchampin> chair: pchampin
14:49:33 <pchampin> agenda: https://lists.w3.org/Archives/Public/public-rdf-star/2021May/0052.html
14:49:33 <agendabot> clear agenda
14:49:33 <agendabot> agenda+ Discussion on the media types
14:49:47 <pchampin> agenda?
14:50:23 <pchampin> pchampin has changed the topic to: RDF-star 14/05 - Agenda: https://lists.w3.org/Archives/Public/public-rdf-star/2021May/0051.html
14:50:27 <pchampin> agenda: https://lists.w3.org/Archives/Public/public-rdf-star/2021May/0051.html
14:50:27 <agendabot> clear agenda
14:50:27 <agendabot> agenda+ Announcements and newcomers
14:50:27 <agendabot> agenda+ Open actions
14:50:27 <agendabot> agenda+ Referential opacity
14:50:27 <agendabot> agenda+ Open-ended discussions
14:50:31 <pchampin> agenda?
14:51:02 <pchampin> agenda+ Discussion on media-types
14:52:43 <AndyS> AndyS has joined #rdf-star
14:54:56 <gkellogg> gkellogg has joined #rdf-star
14:58:44 <TallTed> TallTed has joined #rdf-star
14:59:56 <ora> ora has joined #rdf-star
15:00:02 <james> james has joined #rdf-star
15:00:16 <james> present+
15:00:24 <jbollema> jbollema has joined #rdf-star
15:00:27 <thomas> thomas has joined #rdf-star
15:00:33 <jbollema> present+
15:00:38 <jbollema> hi
15:00:46 <ora> present+
15:01:01 <gkellogg> zakim, who is here?
15:01:01 <Zakim> Present: james, jbollema, ora
15:01:03 <Zakim> On IRC I see thomas, jbollema, james, ora, TallTed, gkellogg, AndyS, RRSAgent, Zakim, pchampin, agendabot, rhiaro
15:01:05 <gkellogg> present+
15:01:16 <thomas> present+
15:01:19 <olaf> olaf has joined #rdf-star
15:01:19 <AndyS> present+
15:01:35 <olaf> present+
15:01:58 <pchampin> present+
15:02:40 <william> william has joined #rdf-star
15:03:49 <pchampin> scribe: olaf
15:03:51 <ora> I will not volunteer as a scribe because I know that Zakim prefers me anyway in the lottery. ;-)
15:04:00 <TallTed> present+
15:04:01 <pchampin> agendum 1
15:04:07 <pchampin> zakim, agendum 1
15:04:07 <Zakim> I don't understand 'agendum 1', pchampin
15:04:22 <pchampin> Topic: Announcement
15:04:33 <olaf> pchampin: no newcomers
15:04:44 <pchampin> q?
15:04:53 <olaf> ... and no announcements
15:04:54 <rivettp> rivettp has joined #rdf-star
15:04:57 <pchampin> Topic: Open actions
15:05:01 <rivettp> present+
15:05:09 <pchampin> https://github.com/w3c/rdf-star/issues?q=is%3Aopen+is%3Aissue+label%3Aaction
15:05:36 <olaf> pchampin: unfortunately, didn't find much time to work on open actions
15:05:53 <olaf> ... about 166, deferred to today
15:06:23 <olaf> ... 168 (charter) no progress because of work on another charter (signatures)
15:06:52 <olaf> ... learned a lot during this process, which may come useful for working on our charter
15:07:19 <olaf> ... action about service descriptions, no progress here either
15:07:46 <olaf> ... we may discuss relevant IRIs to be defined
15:08:05 <olaf> ... which means we need our own namespace
15:08:12 <olaf> ... which may then contain all these IRIs
15:08:47 <gkellogg> q+
15:08:51 <olaf> ... another option may be to leave this open for the potential WG to make a decision
15:08:51 <pchampin> ack gkellogg
15:09:06 <olaf> gkellogg: we should record the issues/direction in sections now
15:09:08 <thomas> q+
15:09:30 <olaf> ... because the start of a WG may need a long time
15:09:38 <olaf> ... and things may become forgotten
15:09:39 <pchampin> ack thomas
15:10:03 <olaf> thomas: about example "has occurrence" - that hasn't been discussed
15:10:21 <olaf> ... better leave it out because it hasn't been discussed
15:10:40 <olaf> ... or define a property for it because it is an important use case
15:10:56 <olaf> pchampin: yes, defining such property may make sense
15:10:58 <pchampin> q?
15:11:16 <olaf> thomas: on the other hand, it's incomplete
15:11:42 <olaf> ... because there would be no way to state where the occurrence is
15:11:49 <pchampin> q?
15:11:57 <olaf> ... or we may add vocabulary for that too
15:12:16 <olaf> pchampin: makes sense, please create an issue
15:12:52 <olaf> ... last action: create outstanding issues list
15:13:10 <olaf> ... did some clean up
15:13:18 <olaf> ... in github
15:13:34 <olaf> ... but didn't remove any open issues in github
15:14:07 <pchampin> https://github.com/w3c/rdf-star/issues/121
15:14:11 <olaf> ... things that really need to be done before the final report: #121
15:14:33 <olaf> ... Andy has text that may be included here
15:14:50 <william> william has joined #rdf-star
15:15:22 <olaf> ...(this text is in the github issue)
15:15:55 <olaf> ... additionally, the intro section needs to be tidied up
15:16:15 <olaf> ... RDF-star vocab may wait
15:16:17 <pchampin> q?
15:16:31 <AndyS> Sounds good.
15:16:45 <AndyS> q+
15:16:45 <olaf> olaf: sounds good for me too
15:17:39 <olaf> pchampin: everyone should look at the open issues in github
15:17:57 <olaf> ... and see whether they find some of these important enough
15:17:59 <pchampin> action: everyone to check in the issues if they believe one of the need to be addressed before "final report"
15:18:25 <pchampin> q?
15:18:51 <pchampin> ack AndyS
15:19:00 <olaf> AndyS: merged a text in the doc
15:19:10 <olaf> ... regarding comparison for SPARQL
15:19:36 <olaf> ... that should be added
15:19:53 <olaf> idea: define it recursively
15:20:04 <jbollema> q+
15:20:22 <pchampin> ack jbollema
15:20:23 <olaf> ... first for S, then P, then O
15:20:37 <olaf> jbollema: regarding #121
15:20:52 <olaf> ... embedded triples should be last in the comparison order
15:21:23 <olaf> ... because it allows for a slight optimization if the data is organized in a specific way
15:21:25 <pchampin> q?
15:21:48 <AndyS> Assign it to me.
15:22:09 <pchampin> action: AndyS to make a PR for issue 121
15:22:16 <james> q+
15:22:21 <pchampin> ack james
15:22:56 <olaf> james: symmetry should not be applied
15:23:08 <olaf> ... (#121)
15:23:14 <rivettp> question: regarding action 1, HOW should we indicate issues that we think should be addressed?
15:23:33 <olaf> AndyS: think james misread
15:23:56 <olaf> ... sameTerms is equals but not the other way around
15:24:27 <olaf> james: will review it again
15:24:30 <pchampin> q?
15:24:54 <olaf> pchampin: the PR will be subject to review as well, of course
15:25:05 <olaf> pchampin: anything else?
15:25:06 <pchampin> q?
15:25:11 <olaf> pchampin: none
15:25:27 <pchampin> Topic: Referential opacity/transperency of embedded triples
15:25:37 <AndyS> james - [[ "A = B" is defaults to "sameTerm(A,B)" as it does for URIs. ]] is only stating SPARQL normal for "=". Can you suggest better text?
15:26:35 <olaf> thomas: wrote some emails
15:26:43 <olaf> ... hope everyone read them
15:26:56 <olaf> ... 5 sec summary may be difficult
15:27:12 <olaf> pchampin: stepped back
15:27:26 <olaf> ... this group is not an official WG
15:28:03 <olaf> ... acting as a chair and editor is tricky here
15:28:36 <olaf> ... became less convinced about the current solution
15:28:47 <olaf> .. although it is still the best option on the table
15:29:06 <olaf> ... it's not constructive to repeat arguments again and again
15:29:24 <pchampin> q?
15:29:30 <olaf> q+
15:29:57 <olaf> pchamin: disappointed that the 2nd strawpoll attracted little participation
15:30:17 <gkellogg> q+
15:30:30 <olaf> ... didn't want to seem biased by invoking a 6-months old strawpoll as an argument
15:30:56 <olaf> ... unfortunately, the new strawpoll has muddied the waters more than cleared it
15:31:15 <olaf> thomas: strawpoll is not binding, right?
15:31:25 <olaf> pchampin: right - we are not a WG
15:31:25 <pchampin> ack olaf
15:31:27 <pchampin> scribe+
15:31:51 <pchampin> olaf: I didn't manage to read the long email, unfortunately.
15:32:30 <pchampin> ... I skimmed it. I am not totally agains moving towards a transparent semantics.
15:33:04 <pchampin> ... But I consider the opaque semantics as a better building block.
15:33:28 <pchampin> ... It allows to add "local transparecy" for dedicated properties.
15:33:38 <pchampin> ... I can't see how it could work the other way.
15:34:13 <pchampin> ... You were not convinced by my arguments; that might have been because I was targetting James' question.
15:35:37 <pchampin> ... It was modelled with SPARQL, but can be modelled otherwise.
15:36:14 <pchampin> thomas: strong feeling against attaching it to specific properties.
15:36:24 <pchampin> ... This is not how the semantic web works at the moment.
15:37:38 <pchampin> ... I understand that SPARQL is not the issue here, but I still find this solution too complex.
15:37:43 <pchampin> ... It is not how people use RDF.
15:38:24 <pchampin> olaf: I would like to see more concrete use cases which you think might not be possible.
15:39:04 <pchampin> thomas: everything does not work. This is not the way RDF-star is adverstised.
15:39:36 <pchampin> ... The SemWeb works in a transparent way. You can replace an identifier with another one.
15:39:54 <pchampin> ... Switching in the middle of a sentence from transparency to opacity is confusing.
15:40:27 <AndyS> q+
15:40:39 <pchampin> ... Better to separate concerns, as proposed by Antoine Zimmermann: using literals for opacity.
15:41:25 <pchampin> olaf: you are making strong claims about how people use the Semantic Web.
15:41:39 <pchampin> ... If someone says that A owl:sameAs B, you don't have to accept it.
15:41:53 <pchampin> q?
15:42:37 <pchampin> ack gkellogg
15:42:50 <olaf> scribe+
15:42:51 <pchampin> gkellogg: I out of my depth on this discussion.
15:43:06 <olaf> gkellogg: Peter had reasonable examples in his email
15:43:28 <olaf> ... such as the one about the Berlin population
15:43:36 <jbollema> q+
15:43:49 <olaf> q+ about Peter's Berlin example
15:44:26 <olaf> pchampin: the problem is related; Peter's example shows something that can be confusing
15:44:32 <olaf> ... and we need to address that
15:44:33 <pchampin> ack AndyS
15:45:04 <olaf> AndyS: Peter's example highlights difference between ..(?)
15:45:17 <olaf> ... choice what to apply are local
15:45:30 <olaf> ... entailment is about adding extra believes
15:45:46 <olaf> ... local ontology comes into play
15:45:50 <pchampin> s/..(?)/denotation and entailment/
15:46:23 <olaf> ... Peter argues from a syntax point of view
15:46:40 <olaf> ... Also, it uses D-entailment
15:47:03 <thomas> q+
15:47:07 <olaf> it uses the intuition that this entailment regime is usually considered as globally agreed upon
15:47:19 <pchampin> ack jbollema
15:47:22 <olaf> ... which is different from other entailment regimes
15:47:45 <olaf> jbollema: my understanding of opacity ...
15:47:53 <olaf> ... it is not permissive
15:48:06 <pchampin> q+
15:48:18 <pchampin> q-
15:48:30 <olaf> ... because a query engine must not use owl:sameAs for embedded triples
15:49:07 <pchampin> transparency is not "allowing", it is "mandating"
15:49:19 <olaf> ... second part, having to preserve syntactic form of embedded triples
15:49:31 <pchampin> q?
15:49:39 <olaf> ... maybe a practical issue for systems
15:49:50 <olaf> ... it may prevent practical optimizations
15:50:10 <pchampin> q?
15:50:25 <olaf> AndyS: that#s what I mentioned in my message
15:50:35 <olaf> ... what you do locally is up to you
15:50:45 <olaf> ... the issue is that the reverse is not possible
15:51:05 <olaf> pchampin: to Jerven's question ...
15:51:24 <olaf> ... semantic transparency is not just allowing it but mandating it
15:51:43 <olaf> ... that's "killing" some use cases
15:51:54 <olaf> ... e.g., provenance
15:52:22 <olaf> ... opacity is not as restrictive
15:52:44 <olaf> ... some things may require additional machinary
15:52:50 <olaf> ... but that's the same in RDF
15:53:12 <olaf> thomas: that's a bad argument because we are not caring only about RDF
15:53:28 <pchampin> ack thomas
15:53:39 <olaf> pchampin: saying that this breaks the SemWeb doesn't make sense either
15:53:55 <olaf> thomas: I think I explained in my emails what exactly is breaking the SemWeb
15:54:17 <olaf> ... it is a feature if you can rely on what you get form something else
15:54:36 <olaf> pchampin: yes but you can add
15:54:43 <olaf> thomas: that#s dangerous
15:56:02 <jbollema> they are dangeling in sparql query behaviour
15:56:36 <AndyS> q+
15:56:38 <olaf> ... provenance example ship1 versus ship2
15:57:00 <pchampin> q?
15:57:16 <pchampin> ack AndyS
15:57:16 <jbollema> << :a :b :c >> ?x ?y  would not be found << :d :b :c >> ?x ?y . even if :a owl:sameAs :d was added to the graph and reasoned about
15:57:45 <olaf> AndyS: You make strong statements
15:58:57 <olaf> ... entailment is local but denotation is global
15:59:09 <pchampin> q?
15:59:47 <olaf> thomas: asking pchampin for the distinction between syntax and interpretation
16:00:25 <william> q+
16:01:12 <olaf> ... the problem wouldn't exist if we had an explicit identifier for the embedded triple
16:01:24 <pchampin> q?
16:02:58 <olaf> pchampin: it really depends on what you want to identify with an embedded triple -- a triple or a statement
16:03:24 <pchampin> ack william
16:03:34 <olaf> ... the triple itself is closer to the syntax and allows for more use cases
16:04:06 <olaf> william: in N3 we have formulas which have ref.opacity
16:04:16 <olaf> for quoted graphs
16:04:35 <olaf> ... the reasons for choosing it this way is because it allows to add more things on top
16:04:50 <thomas> but its a different use case
16:04:53 <olaf> ... in contrast to having ref.transparency
16:05:03 <thomas> because they stand for themselves
16:05:32 <olaf> ... interpretation of cited formula may differ
16:06:12 <gkellogg> That’s a pretty convincing argument for defaul opacity.
16:06:17 <olaf> ... ref.opacity is the building block that allows for this
16:06:57 <pchampin> STRAWPOLL: do we keep a referentially opaque?
16:07:08 <gkellogg> +1
16:07:11 <william> +1
16:07:12 <olaf> +1
16:07:15 <pchampin> s/opaque/opacity/
16:07:16 <pchampin> +1
16:07:17 <thomas> -1
16:07:19 <AndyS> +0.9
16:07:25 <james> -1
16:07:27 <ora> 0
16:07:36 <TallTed> +0.5
16:07:49 <TallTed> s/referentially/referential/
16:07:58 <jbollema> -1
16:08:17 <william> for those who are interested: http://ceur-ws.org/Vol-2438/paper6.pdf
16:08:40 <olaf> pchampin: still hardly a consensus ... unfortunately
16:09:13 <TallTed> q+
16:09:19 <olaf> ... we have to reflect this lack of consensus in the document
16:09:38 <ora> Good discussion
16:09:55 <olaf> ... other item on the agenda (mimetypes as propsoed by James') postponed to next week
16:10:20 <william> thanks!
16:10:34 <jbollema> bye
16:10:44 <olaf> Bye!
16:10:49 <olaf> olaf has left #rdf-star
18:00:58 <gkellogg> gkellogg has joined #rdf-star
20:04:30 <gkellogg> gkellogg has joined #rdf-star
21:16:44 <gkellogg> gkellogg has joined #rdf-star
23:27:32 <gkellogg> gkellogg has joined #rdf-star