RIF Telecon 1-Sept-09 -- 01 Sep 2009

admin

<ChrisW> Scribe: LeoraMorgenstern

Liaison

sandro, chris: OWL, like RIF, relies on XSD 1.1., and that is still in candidate recommendation.

sandro: prognosis for XML schema datatypes is not promising.
... OWL will have an appendix referring to XML schema datatypes.
... and RIF can do something similar.

<ChrisW> next item

Nothing else in liaison.

<ChrisW> next item

<sandro> ACTION: sandro to make sure if OWL does a normative appendix for XSD 1.1, that it's phrased in a way that makes it also work for RIF. [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action01]

<trackbot> Created ACTION-903 - Make sure if OWL does a normative appendix for XSD 1.1, that it's phrased in a way that makes it also work for RIF. [on Sandro Hawke - due 2009-09-08].

Action Review

<ChrisW> close action-902

<trackbot> ACTION-902 Look at message from Nick B. and check FLD schema closed

close action-898

<trackbot> ACTION-898 Answer faq 3.9 closed

Exit Criteria

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

<johnhall> zakim ??p0 is me

<johnhall> P0

<johnhall> zakim P0 is me

<johnhall> Thanks Jos

Chris: The exit criteria for RIF are listed at http://www.w3.org/2005/rules/wiki/Exit_Criteria

Sandro: I am concerned that Chris's revision to my proposal --- namely, reducing the requirement from implementing a dialect of FLD to merely specifying the dialect --- is putting the bar too low.
... I don't think it's too high a burden to require an implementation.

Chris: I think it's an unnecessary burden.
... Regarding Sandro's point that someone can just flip through FLD and figure out how to have some sort of (trivial?) instantiation: Someone can also just say that they have an implementation.

Sandro: Shouldn't they at least have to show that they can read and write XML?

Chris: Implementation tests the dialect, not FLD.

Sandro: I won't object; just wanted to make the point.

Harold: Should we also require a syntax and semantics for the dialect specification?

Chris thereupon made that change to the Exit Criteria.

<ChrisW> PROPOSED: accept http://www.w3.org/2005/rules/wiki/index.php?title=Exit_Criteria&oldid=10799 as RIF CR Exit criteria

<josb> +1

<StellaMitchell> +1

<Harold> +1

<ChrisW> +1

<sandro> +1

<LeoraMorgenstern> +1

<johnhall> +1

<ChrisW> RESOLVED: accept http://www.w3.org/2005/rules/wiki/index.php?title=Exit_Criteria&oldid=10799 as RIF CR Exit criteria

Publications

Chris: Editors need to make changes to their Last Call documents: BLD, PRD, DTB, SWC, FLD, Core
... Needs to be done by next week's meeting.
... Assignments: Jos, SWC; Chris, DTB; Christian, PRD; Harold, BLD, FLD, Core.

Sandro: Even if no changes have been made, one needs to note in the document that there have been no changes since Last Call draft.

<sandro> target publication date: Setp 17

<sandro> try to approve WD pubs of Test and UCR on Sept 15.

<sandro> PROPOSED: Our Last Call drafts (Core, BLD, PRD, DTB, SWC, and FLD) are ready to be published as Candidate Recommendations

<ChrisW> +1

<sandro> +1 (W3C)

<ChrisW> (IBM)

<Harold> +1 (NRC)

<josb> +1 (FUB)

<johnhall> +1 (OMG)

<StellaMitchell> +1 (self)

<LeoraMorgenstern> +1 (self)

<sandro> RESOLVED: Our Last Call drafts (Core, BLD, PRD, DTB, SWC, and FLD) are ready to be published as Candidate Recommendations

Sandro: Note that the statement of no changes or changes to last call drafts is CRITICAL PATH, and therefore should be done today or tomorrow.

<sandro> "Change since the 3 July draft...." or "Changes since the Second Last Call draft of 3 July..."

<josb> tomorrow is fine for SWC changes statement

Implementations

Chris: Since we have moved last call documents to candidate recommendation, we are now in the call for implementations period.
... We had originally talked about a two-month period for implementations.
... To get out of CR, we need implementations.

Sandro: Yes, those are in the exit criteria.

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

Sandro: we want to point people to the above wiki page.

Chris: Now is the time to follow up with people who indicated in the comments that they would be interested in providing an implementation.
... We need to get commitments from these people that they will do this in this period.
... Let's begin by making a list of people who will do implementations.

csma: ILOG is working on an implementation. A month ago, we had a first prototype running, so we probably have more going on now.

sandro: I don't know if I'll get one done.

Chris: Someone in the XML group at IBM made a RIF-based rule storage system. Really a demo of XML X-query technology, but they did take some rules and translated them to RIF, and stored them in RIF-XML, and queried them.
... question: does that count as a RIF implementation?

<ChrisW> ACTION: try to dig up XML RIF store [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action02]

<trackbot> Sorry, couldn't find user - try

Sandro: Mike Dean has an implementation.

<ChrisW> ACTION: Chris to try to dig up XML RIF store [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action03]

<trackbot> Created ACTION-904 - Try to dig up XML RIF store [on Christopher Welty - due 2009-09-08].

mdean: RIF implementation is in progress.

Chris: Silk will be an implementation of BLD.

csma: Gary mentioned he was doing something, but I don't know the status of it now.
... will send Gary a message, asking for status.

<ChrisW> ACTION: Christian to ask Gary about status of implementation [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action04]

<trackbot> Created ACTION-905 - Ask Gary about status of implementation [on Christian de Sainte Marie - due 2009-09-08].

csma: I think Adrian is working with tibco on an implementation of PRD.

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

Sandro: maybe we should have a RIF Dev mailing list, similar to OWL dev?

<sandro> ACTION: sandro request creation of public-rif-dev [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action05]

<trackbot> Created ACTION-906 - Request creation of public-rif-dev [on Sandro Hawke - due 2009-09-08].

Chris: Rolf Gruetter has said that his group at WSL needs to have disjunction in rule heads.

Note: Gruetter did not say anything about his group actually developing an implementation

csma: OntoBroker has some sort of implementation of BLD

Chris: Alexander Riazanov working on an implementation that converts BLD to TPTP.

Harold: He has been on vacation; but I'll try to talk to him about that.
... There needs to be some web page on implementation of OntoBroker

Sandro: Or better yet, send in an implementation report, as my wiki page specifies.
... Doing that is very quick.

Chris: Currently the public information about OntoBroker's implementation is inconsistent. ontoprise web page says there's a RIF implementation and links to OntoBroker web page, but OntoBroker's web page says nothing.
... What about Tom Gordon. Is he an implementor?

Harold: I don't think he's an implementor. He has a system LKIF, which is for legal knowledge.
... I don't believe he will implement RIF: he has more of a theoretical interest in whether one can represent legal knowledge using RIF.

<sandro> Chime

<sandro> Chimezie Ogbuji

Discussion also on Adreas Abecker's comments and Chimezie Ogbuji's comments.

<ChrisW> Nick Bassiliades

Harold: Nick Bassiliades has been following RIF; does defeasible rules; unclear as to whether he'll actually do an implementation of RIF.

<sandro> CR dreadline Oct 23rd, friday before the conferences....

Sandro: Let's make deadline for implementations October 23, so people will be able to announce it before the rules conferences.

Chris: We'll revisit this topic at each telecon.

csma: will restart work on RIF XML soon.

Test Cases

Chris: Axel had made an all built-ins test case
... And that test case seems to indeed include all built-ins

Stella: I think it would be better to split it up somehow.

Chris: Perhaps by data-type?
... There are definitely some typos, like "listeral" instead of "literal"

Stella: Perhaps organize it by string predicates, number predicates, etc?

<josb> (I wanted to say exactly what Chris just said)

Sandro: what's the problem with it being so big?

Stella: If it fails, it's hard to figure out why.
... can we split it by positive guards, negative guards, etc?

Sandro: We can group it by the things people are most likely to implement

<ChrisW> ACTION: Stella to refactor All Builtins testcase [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action06]

<trackbot> Created ACTION-907 - Refactor All Builtins testcase [on Stella Mitchell - due 2009-09-08].

Stella: I can refactor the all built-ins case
... There are all sorts of issues still to be dealt with, with respect to the test cases.
... Especially with all the changes in the documents, the specifications, etc.
... We need more than 15 minutes to go through this.

<StellaMitchell> http://lists.w3.org/Archives/Public/public-rif-wg/2009Jul/0026.html

Chris: Let's make this a priority for the next telecon, since the set of test cases will be very important to the implementors.
... What do we do with unapproved test cases? Require changes? Drop them?
... We do need an Assert/Retract case

csma: We need an Assert /Retract case that is different from a Modify.
... I don't have a case in mind, but it probably should be a negative case; that you can't assert something about an object that you've retracted.
... I had a discussion with Adrian about this. He initially didn't agree; then we had a discussion, including Changhai Ke and Gary, who agreed with me. But Adrian has not replied, and therefore there has been no conclusion.

<ChrisW> ACTION: Christian to fix/update AssertRetract test case [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action07]

<trackbot> Created ACTION-908 - Fix/update AssertRetract test case [on Christian de Sainte Marie - due 2009-09-08].

csma: I need to get a consensus on this, and then either modify the test case myself, or get someone to do it.
... I think the assert test case is fine.
... However, there is no XML for it.

Chris: do we have anything to generate XML for PRD?

csma: Not yet.

<StellaMitchell> jacc

Stella: can the XML be generated automatically using a tool like jacc?

Chris: What about the other PRD test cases? Won't this be a problem for all of them?

<ChrisW> ACTION: Christian to check into XML syntax for PRD test cases [recorded in http://www.w3.org/2009/09/01-rif-minutes.html#action08]

<trackbot> Created ACTION-909 - Check into XML syntax for PRD test cases [on Christian de Sainte Marie - due 2009-09-08].

Chris: 2 remaining open issues in Working Group:

<ChrisW> PROPOSED: close issue-37 as it is addressed by the draft note on RIF combination with XML Data

<StellaMitchell> +1

<ChrisW> +1

<johnhall> +1

<csma> +1

<sandro> +1

<josb> +1

<mdean> +1

<ChrisW> RESOLVED: close issue-37 as it is addressed by the draft note on RIF combination with XML Data

<ChrisW> PROPOSED: close issue-38 as it is addressed by the draft note on RIF combination with XML Data

<ChrisW> +1

<csma> +1

<sandro> +1

<mdean> +1

<johnhall> +1

<LeoraMorgenstern> +1

<josb> +1

<ChrisW> RESOLVED: close issue-38 as it is addressed by the draft note on RIF combination with XML Data

<StellaMitchell> +1

<ChrisW> NEXT MEETING IN TWO WEEKS!

RIF Telecon 1-Sept-09

01 Sep 2009

Attendees

Contents

admin

Liaison

Action Review

Exit Criteria

Publications

Implementations

Test Cases

Summary of Action Items

Scribe.perl diagnostic output