17:01:55 RRSAgent has joined #rdfn-s 17:01:55 logging to http://www.w3.org/2010/06/27-rdfn-s-irc 17:02:04 Anchakor has joined #rdfn-s 17:02:12 ekw has joined #rdfn-s 17:02:17 RRSAgent, make logs public 17:02:33 Semantics break out 17:02:50 scribenick: jjc 17:02:58 pfps has changed the topic to: RDF Next Steps Workshop, Semantics breakout (pfps) 17:03:04 Topic: Semantics 17:03:55 Four issues at 17:04:02 http://www.w3.org/2001/sw/wiki/RDF_Core_Charter_2010#Semantics 17:04:29 Blank Nodes, SPARQL-Style RDF Semantics 17:04:44 Archaisms, Literals as Subjects 17:04:57 Jeremy and Peter do not like SPARQL-Style RDF Semantics 17:05:12 as a title 17:07:13 Evan: let's start with the easy bits 17:07:31 Paul: rdf:value - keep this to deal with literals as subjects 17:09:46 List of Archaisms under syntax: 17:09:54 * reification (align with some "named graph" support) 17:09:54 * rdf:Alt (just remove) 17:09:54 * rdf:Bag (fix or remove) 17:09:54 * rdf:Seq (use rdf:List instead; it's costly to have two similar options) 17:09:54 * rdf:value 17:09:55 * xs:string (use plain literals instead) 17:09:57 * rdf:XMLLiteral (use plain literals instead, with quoted XML) 17:09:59 * rdf:ID (use rdf:about instead) 17:11:51 What should we do for these? 17:12:41 discussion of Archaisms 17:14:53 reification: Jeremy - there is no semantics 17:15:22 containers: Jeremy there is no semantics 17:15:33 Peter: yes there is container membership 17:15:37 Jeremy: yes 17:15:43 Paul: what do we want to do 17:16:41 Peter: no one knows what containers mean 17:16:55 Paul: I am reluctant to deprecate something which we do not have equivalent 17:17:06 Peter: we do, you can roll your own 17:17:52 Jeremy: if we are going to depreacte anything, we deprecate Alt 17:18:05 James: let's add to the list for discussion 17:19:59 Peter: we could remove axiomatic triples for rdf:_nn 17:20:47 and hence all the semantics 17:22:01 Paul: are we happy with the semantics of xs:string and plain literals being different 17:22:14 Jeremy: they're not - most widely misimplemented part of RDF Semantics 17:22:43 Peter: simple and D entailments to do with these string literals are different 17:23:06 XMLLiteral .... 17:24:09 rathole into whether we can have inconsistency of RDF or is it RDFS 17:25:51 In RDFS you can get an inconsisteny with an ill-typed XML Literal 17:26:02 in D semantics also illformed xsd literals 17:26:14 ACTION: pfps edit wiki to include list of Archaism 17:27:53 Jeremy: does anyone else have other Archaims 17:28:12 e.g. ill-formed lists 17:28:23 Peter: this would be a big change to the semantics 17:28:28 you could infer sameAs 17:29:31 brief discussion of rdfs:comment etc. 17:30:11 topic: blank nodes 17:30:24 James: does mulgara change the labels internally 17:30:32 Paul: no -we reserve the right to 17:31:23 Jeremy: I don't like the wording 17:31:28 James: is it about scope 17:33:22 Paul: qu about SPARQL and b node semantics 17:34:32 Ian apparently claims that current semantics of blank nodes creates NP complete problem in SPARQL 17:34:48 Jeremy: but subgraph isomporphism is there anyway because of variables 17:34:53 and this is NP complete 17:34:55 Peter: yes 17:35:05 Paul: but we have b node semantics documented 17:35:07 but not used 17:41:17 Peter: maybe it was a mistake to talk about entailment 17:41:25 then we talk about query instead 17:41:53 But maybe this becomes a SPARQL issue 17:42:02 that they should not talk about entailment 17:42:51 --- BREAK 17:43:19 Before break: Jeremy we are not progressing 17:43:31 Evan: but you seem to be making progress on entailment issue 17:59:52 ekw has joined #rdfn-s 18:04:34 scribenick: ekw 18:06:00 paul: we should look at blank nodes and decide if they are addressable 18:06:29 ...put it forward as a proposal or say its not something we can do 18:06:54 alejandro: people are using blank nodes in different ways 18:06:55 we need to remember to put in stuff to fix the inference rules 18:07:07 ... why can DBpedia create a huge 18:07:19 DB without them, while others can't 18:08:45 jjc: people think that they shouldn't use them, but then they misuse other features to do the same thing 18:09:26 jjc: I think that we have agreed that the formal semantics of RDF will change, but the informal semantics should preserve the blank node disctinction 18:10:24 alehandro: do we need a suggested way of handling this (the pragmatic support for blank nodes) 18:10:38 pfps: there are two things that we could do 18:11:05 ... we could decide that blank nodes as currently spec'd are just bad 18:11:28 ... you might just decide "There are no blank nodes in RDF graphs." 18:12:02 jjc: a blank node is a node who's name that we don't know rather than a node without a name (an existential) 18:12:25 pfps: the semantics wouldn't have to change for this 18:13:03 jjc: we could delete the text describing blank nodes as an existential since people don't understand it anyway 18:14:14 paul: if we go to SemTech and talk with org's using this stuff, how are they looking at blank nodes? 18:14:41 jjc: the smallest change would be to provide a skolemization algorithm for blank nodes 18:16:01 ... however that wouldn't address the issue of the folks who are against using blanknodes 18:16:37 paul: this could change the ways graphs merge 18:17:00 ... I don't know if it breaks what's out there, but it certainly changes the story. 18:22:42 jjc: another con is it may make it harder to compare 2 graphs using graph isomorphism 18:25:22 paul: we have established that blank nodes could benefit from some work 18:28:38 paul: is it possible with working on blank nodes that it could change the semantics? If so, we should record this. 18:29:51 jjc: should we look at literals as subjects now? 18:30:37 PFPS not JJC on last comment 18:32:23 paul:I can't think of a use case for a literal as a predicate, but can for a literal as a subject. 18:32:56 ... 3 squareRootOf 9 for example 18:33:42 Stephan: What's the danger of this? 18:34:09 jjc: A con is that rdf/xml changes would be significant. 18:35:30 jjc: as pro it would actually simplify some situations. 18:36:15 jjc: on OWL systems, its difficult but in a triple store system it's significantly easier. 18:39:24 Some discussion about how this relates to the restriction on InverseFuntional properties in OWL. 18:42:13 Going forward this would add more divergence between RDF and OWL (causing a document fork) 18:42:40 Stephan: I don't find these use cases compelling enough to justify this change. 18:42:55 ... what I have mind is the social impact that this would have. 18:44:08 This could cause people to make statements about the string instead of about the entity. 18:45:32 Paul: lack of this feature has annoyed me. 18:45:53 ... such as in data bases of pseudo primes. 18:48:48 jjc: I don't think we should be neutral about it. I think this group should say that this feature is not worth doing. 18:49:48 jjc: although one of the things that TBL wants is graph literals as subjects. 18:50:45 Paul: how about a best practice of making a blank node sameAs a literal, and use the blank node as a subject everywhere? 18:51:20 Stephan: That would be to me, shooting yourself in the foot. 18:51:56 Stephan: I feel unhappy about these type of proposal. 18:53:21 ... what's wrong with suggesting that a group investigate this and collect use cases? 19:07:00 subtopic: SPARQL-style semantics 19:10:42 jjc: I would suggest that introductory blurb should talk about RDF as a data language rather than "as seen in SPARQL systems" 19:31:20 [scribe had to leave the room for a half hour] 20:33:10 pfps has joined #rdfn-s 20:33:43 jjc has joined #rdfn-s 20:34:26 pgearon has joined #rdfn-s 20:38:47 ekw has joined #rdfn-s 20:39:06 AndyS has joined #rdfn-s 20:39:53 Issues of simplified RDF semantics 20:41:41 AndyS1 has joined #rdfn-s 20:42:57 Scribe: Andy Seaborne 20:43:01 scribenick: AndyS 20:50:54 Section: Semantics for Next Steps 20:51:21 ... datasets: alternatives appraoches for semantis need to be 20:52:03 ... considered including model theoretic or (if now normative) entailment rules 20:53:14 Are named graphs first class citizens? Both "yes" and "no" are possible. 20:54:49 JJC: Q: what is the relationship of two named graphs in a single document? If any (leave pragmatic)? 20:55:49 PFPS: When synatx is done, not all work is done. 20:56:42 -> require that any extensions come with semantics 21:00:01 Section: Blank Nodes 21:00:18 timbl has joined #rdfn-s 21:02:04 add possibility of global naming for blank nodes (not URIs , maybe URIs - alternatives) 21:03:22 Issue: currently reading twice creates different bnodes - preserve this? 21:03:42 JJC: (round tripping discussion) 21:04:14 ... want to round trip i.e. preserver references, across serialization and exchange. 21:04:41 ... e.g. "this bnodes (triples using...) have changed" 21:06:08 Blank nodes ... remote references use cases ... 21:06:46 ... skolemized: what is not yet skolemized? 21:07:00 skolemization is not bnode label 21:08:22 ... editting may not preserve skolemization (large, ugly, not human friendly) 21:10:11 ... speculation on possible solutions 21:10:29 ... to scope space of possibilities 21:10:46 rrsagent, make logs public 21:22:31 editting the blank nodes section 21:24:40 End of break out session 21:31:14 rrsagent, make minutes 21:31:14 I have made the request to generate http://www.w3.org/2010/06/27-rdfn-s-minutes.html pfps 21:31:46 jeremycarroll has joined #rdfn-s 21:39:53 pgearon has joined #rdfn-s 21:40:06 AndyS has joined #rdfn-s 21:41:49 ekw has joined #rdfn-s 21:47:06 pgearon has left #rdfn-s 21:56:42 AndyS1 has joined #rdfn-s 22:31:59 AndyS has joined #rdfn-s 22:33:26 ekw has joined #rdfn-s 22:43:03 AndyS has left #rdfn-s 22:51:44 ekw has joined #rdfn-s 23:34:55 pfps has left #rdfn-s 23:53:06 ekw has joined #rdfn-s