RIF Telecon 13-Oct-09

 

<AxelPolleres> I'd have some AOB, but can only dial in later: http://lists.w3.org/Archives/Public/public-rdf-dawg/2009OctDec/0084.html, http://lists.w3.org/Archives/Public/public-rif-wg/2009Oct/0017.html

<ChrisW> Scribe: MichaelKifer

<ChrisW> MInutes from last telecon: http://lists.w3.org/Archives/Public/public-rif-wg/2009Sep/att-0118/RIF-minutes-29-sep-2009.html

<ChrisW> PROPOSED: accept minutes of last telecon

<ChrisW> RESOLVED: accept minutes of last telecon

<ChrisW> axel, remind us when you join about your OB

Liason

<sandro> http://www.w3.org/2001/sw/rdb2rdf/

Chris: new WG called RDB2RDF (mapping relational DBs to RDF/OWL)

Sandro: Someone from our group should join RDB2RDF to show them the light.

<DaveReynolds> Charter says: "The mapping language SHOULD use W3C RIF whenever a rule engine is needed in the mapping language"

Cris: perhaps BLD or some small extension can do all what RDB2RDF might ever want.

action review

<ChrisW> close action-904

<trackbot> ACTION-904 Try to dig up XML RIF store closed

Public comments

<ChrisW> http://www.w3.org/2005/rules/wiki/Response_to_WL2

<ChrisW> ACTION: Chris to send http://www.w3.org/2005/rules/wiki/Response_to_WL2 [recorded in http://www.w3.org/2009/10/13-rif-minutes.html#action01]

<trackbot> Created ACTION-929 - Send http://www.w3.org/2005/rules/wiki/Response_to_WL2 [on Christopher Welty - due 2009-10-20].

<ChrisW> http://lists.w3.org/Archives/Public/public-rif-comments/2009Sep/0006.html

<ChrisW> http://lists.w3.org/Archives/Public/public-rif-comments/2009Sep/0009.html

<ChrisW> ACTION: michael to look at http://lists.w3.org/Archives/Public/public-rif-comments/2009Sep/0009.html [recorded in http://www.w3.org/2009/10/13-rif-minutes.html#action02]

<trackbot> Sorry, amibiguous username (more than one match) - michael

<trackbot> Try using a different identifier, such as family name or username (eg. msintek, mkifer, merdmann)

<ChrisW> ACTION: mkifer to look at http://lists.w3.org/Archives/Public/public-rif-comments/2009Sep/0009.html [recorded in http://www.w3.org/2009/10/13-rif-minutes.html#action03]

<trackbot> Created ACTION-930 - Look at http://lists.w3.org/Archives/Public/public-rif-comments/2009Sep/0009.html [on Michael Kifer - due 2009-10-20].

Publicity

<Harold> Alexandre Riazanov just wrote:

<Harold> 1. A converter from RIF BLD to the TPTP format. This will allow

<Harold> using all the TPTP reasoners. I already have a working implementation,

<Harold> although it does not deal will External terms and atoms yet.

<Harold> 2. Support for the new TPTP format in VampirePrime. It already works

<Harold> for proof search, and soon I will restore the support of query answering.

<Harold> In particular, the generation of schematic answers over RDB will be supported.

<Harold> http://www.cs.miami.edu/~tptp/

Implementation of BLD: Ontoprise, BBN

An initial FLD dialect for SILK at the end of October.

<AxelPolleres> 'd have some AOB, but can only dial in later: http://lists.w3.org/Archives/Public/public-rdf-dawg/2009OctDec/0084.html, http://lists.w3.org/Archives/Public/public-rif-wg/2009Oct/0017.html

SPARQL licason

2nd FLD dialect from Stony Brook: DLD with ASP semantics.

<AxelPolleres> At the moment, we use:

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#Simple

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#RDF

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#RDFS

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#D

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#OWL-DL

<AxelPolleres> http://www.w3.org/2007/rif-import-profile#OWL-Full

Test Cases

Review of RDF combination subclass test cases

Discussion of what it means to pass a negative entailment case.

Chris: passing a negative test case means failure to derive.

<LeoraMorgenstern> q

Stella: non-passing any of the test case = non-conformance

<APaschke> otherwise the test cases would not make sense

<StellaMitchell> A conformant RIF consumer should report that the conclusion is not entailed by the premises, should not report that the answer is undecided, and must not report that the conclusion is entailed by the premises.

Passing all of them != conformance -- just means the implementation is "looking good".

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass

<APaschke> +1

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass

<mdean> +1

<sandro> sandro: that "should not" is kind of hard to understand. I'd prefer something like, "If no answer is determined, UNDECIDED" is the right answer. One SHOULD try to determine an answer."

By the way, no engine cat always tell that something does not follow, since BLD is only semi-decidable.

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_2

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_2

<StellaMitchell> +1

<APaschke> +1

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_2

<AxelPolleres> any reason why the non-conclusion is red?

<AxelPolleres> ... in http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_3

<AxelPolleres> ?

<StellaMitchell> to draw attention to the fact that it is not entailed

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_3

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_3

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_4

MK: unrelated to these test cases, we need to resolve the semantic problems with the RDF/OWL combination, which was discussed in September.

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_4

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_4

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_5

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_5

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_6

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/RDF_Combination_SubClass_6

<ChrisW> close action-848

<trackbot> ACTION-848 Review RDF Combination subclass test cases closed

<ChrisW> http://www.w3.org/2005/rules/wiki/NestedListsAreNotFlatLists

<ChrisW> PROPOSED: http://www.w3.org/2005/rules/wiki/NestedListsAreNotFlatLists

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/NestedListsAreNotFlatLists

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/NestedListsAreNotFlatLists

<ChrisW> http://www.w3.org/2005/rules/wiki/NotAssertRetract

<ChrisW> close action-849

<trackbot> ACTION-849 Review NestedLists and NotAssertRetract test cases closed

<APaschke> yes, it is obsolte

<sandro> http://www.w3.org/2005/rules/wiki/MalformedLists

<Harold> The terminology was changed: http://www.w3.org/2005/rules/wiki/BLD#Terms. 4. List terms.

<Harold> "Open lists are usually written using the following: List(t1 ... tm | t)."

<sandro> renamed to http://www.w3.org/2005/rules/wiki/OpenLists

<ChrisW> PROPOSED: Accept http://www.w3.org/2005/rules/wiki/OpenLists

<sandro> +1

<ChrisW> RESOLVED: Accept http://www.w3.org/2005/rules/wiki/OpenLists

<ChrisW> http://www.w3.org/2005/rules/wiki/Modify

<ChrisW> PROPOSED: Accept http://www.w3.org/2005/rules/wiki/Modify

<sandro> +1

<ChrisW> RESOLVED: Accept http://www.w3.org/2005/rules/wiki/Modify

<ChrisW> http://www.w3.org/2005/rules/wiki/Modify_loop

<ChrisW> PROPOSED: Accept http://www.w3.org/2005/rules/wiki/Modify_loop

<sandro> +1

<ChrisW> RESOLVED: Accept http://www.w3.org/2005/rules/wiki/Modify_loop

<APaschke> PRD presentation syntax

<APaschke> http://www.w3.org/2005/rules/wiki/PRD#Presentation_syntax_.28Informative.29

<sandro> DaveReynolds: Uhh, shouldn't this have Externals ?

<APaschke> FORMULA ::= IRIMETA? 'External' '(' Atom ')'

<APaschke> TERM ::= IRIMETA? (Const | Var | List | 'External' '(' Expr ')')

<AxelPolleres> so this is not an external function here?

Gary: external is optional in terms

<AxelPolleres> confused

<APaschke> presentation syntax of PRD is in section 9

<AxelPolleres> and there is still: Name ::= UNICODESTRING

<ChrisW> ACTION: stella to send CSMA fixes for PRD EBNF [recorded in http://www.w3.org/2009/10/13-rif-minutes.html#action04]

<trackbot> Created ACTION-931 - Send CSMA fixes for PRD EBNF [on Stella Mitchell - due 2009-10-20].

<AxelPolleres> christian might want to have a look at the recent EBNF changes in FLD/BLD/DTB in that respect as well...

MK: EBNF says external is mandatory, which clashes with the text. But EBNF is only informative.

<AxelPolleres> even if only informative, guess we'd prefer to have it consitent ;-)

<sandro> http://www.w3.org/2005/rules/wiki/Modify_noloop

<Harold> I think we should not change External(...) being mandatory at this stage. (MK: Harold meant "non-mandatory", i.e., to keep it mandatory.)

<Harold> (This will also help to keep BLD and PRD as compatible as possible.)

MK: But in BLD External is already mandatory for external functions and builtins. (MK was confused by Harold's mistyping "mandatory" instead of "non-mandatory")

<Harold> Yes, thats why it should also stay mandatory for external functions and builtins in PRD.

<ChrisW> PROPOSED: Accept http://www.w3.org/2005/rules/wiki/Modify_noloop

<ChrisW> RESOLVED: Accept http://www.w3.org/2005/rules/wiki/Modify_noloop