W3C

- DRAFT -

RIF WG Telcon

11 Jul 2006

See also: IRC log

Attendees

Present
AlexKozlenkov Allen_Ginsberg AxelPolleres CSMA DaveReynolds DavidHirtle Elisa_Kendall EvanWallace FrankMcCabe Gary_Hallmark GiorgosStoilos HaroldB IanH JeffP Leora_Morgenstern MichaelKifer MoZ PaulaP SaidTabet StellaMitchell daveH daver igor johnhall josb pfps sandro
Regrets
Chair
Christian St Marie
Scribe
FrankMcCabe

Contents


<csma> topic 'RIF 11 July http://lists.w3.org/Archives/Public/public-rif-wg/2006Jul/0013.html'

CSMA:Shall we continue meetings as normal during the summer

<AxelPolleres> +1 normally

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2006Jul/att-0018/minutes.txt

csma: Propose to accept above minutes

<Harold> +1

<PaulaP> +1

<AlexKozlenkov> +1

csma: Minutes of july 4th accepted

<AxelPolleres> DaveR sent a request to modify the minutes

<JeffP> It is addressed.

pfps: No change for F2F4

<AxelPolleres> http://lists.w3.org/Archives/Public/public-rif-wg/2006Jul/0015

<AxelPolleres> ok

<AxelPolleres> http://www.w3.org/2005/rules/wg/track/actions/28

<AxelPolleres> ACTION: 28 CLOSED [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action01]

csma:There are no reports regarding liaisons with SPARQL, XQUERY, XPATH, PRR, SBVR

Elisa: ODM was recommended for adoption at OMG Boston Meeting
... Finalization Task Force starts in September

<AxelPolleres> Can you send a pointer to these discussions?

DaveR What is our position on the Semantic Coordination Group

csma: The rule layer in SW stack needs to be synched with RIF
... Is the RIF going to develop a rule language or is a rule interchange format different to a rule language
... This needs to be clarified with the SWCG

daver: debate on fourth goal for RIF to develop rule language not reflected on SWCG

csma: I wrote a personal opinion that it was not a goal, not part of the charter
... no decision, currently under dicsussion

daver: I see the followup message, that it was your opinion not the RIF's policy
... Is this an extension of the charter or not

csma: I prefer to interpret the charter in a minimal way given that it is already v. broad in scope
... Jury out on whether RIF implies rule language or not

daver: Followup in the SWCG following any discussion/decision in RIF

<JeffP> ACTION: csma will let SWCG know that there is an ongoing discussion and they will be updated [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action02]

<Zakim> DavidHirtle, you wanted to ask sandro re: possibility of edit

daveH: Style of UCR doesnt work well in IE or Firefox

<scribe> ACTION: sandro: to fix the styles of the UCR [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action03]

<DavidHirtle> thanks Sandro; I wasn't sure if anything could be done after publication

csma: How do we get feedback.

<AlexKozlenkov> I think we should contact commercial most rule vendors

csma: The liaisons should spread the word in their respective groups

<AlexKozlenkov> Yes, I could do it

<scribe> ACTION: AlexKozlenkov will contact commercial rules vendors to take a look at the UCR [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action04]

<josb> I could notify the SPARQL wg again, since we have no "real" liaison at SPARQL

AlexK: there are now many rules languages/vendors. Ensuring interoperability will be hard.

<scribe> ACTION: AlexK will contact JSR94 community [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action05]

csma: action 33 closed
... action 38 continued
... action 9 continued
... action 41 continued

<PaulaP> continued

csma: action 39 continued
... action 35 closed
... action 55 closed

csma: action 52 closed

<AxelPolleres> ;-)

<AxelPolleres> yes.

<AxelPolleres> ;-)

csma: action 53 continued

<DavidHirtle> done

<DavidHirtle> http://www.w3.org/2005/rules/wg/wiki/UCR/Document_issues

<josb> Notification of the publication of the 2nd WD of the UCR document has been sent to DAWG

csma: action 56 closed
... issues against UCR document should be recorded as issues

<scribe> ACTION: Sandro rename use and requirements into one category [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action06]

csma: We need to prioritize issues
... discuss this next meeting

<scribe> ACTION: csma put organization of issues on next agenda [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action07]

RIFRAF

Michael: Cannot remember action 44
... Cannot remember action 43

HaroldB: How do languages supprt built-ins?
... There was some confusion about the status of the last powerpoint slides

cma: Yes.

csma: Many of the requirements were moved to the RIFRAF
... Do we have the proper criteria in the RIFRAF
... E.g. RIF should support production rules, first order, normative rules, combined language rule sets

<EvanWallace> +1 to not putting RIF requirements in RIFRAF

csma: RIFRAF contains discriminators not requirements

<DaveReynolds> RIFRAF gives a framework in which to define the space of coverage which we require RIF to cover

csma: We need enough discriminators to cover the field

<AlexKozlenkov> Absolutely, Phase 2 requirements should not simply disappear

csma: Group languages into related groups and derive requirements from the groups

haroldb: does space mean that not everything needs to be covered

daver: framework to allow us to describe space of languages. Some will be covered, some not

csma: The choices for covered languages depend on the criteria themselves
... We might not see e.g. production rules emerge without the right discriminators e.g. actions on conclusions
... How do we enusre coverage of discriminators

<AlexKozlenkov> There is currently not enough detail for production or reaction rules.

<csma> http://www.w3.org/2005/rules/wg/wiki/UCR/Critical_Factors_Analysis

There are coverage requirements in the CFA analysis

<PaulaP> ok

<PaulaP> I'll try :)

<scribe> ACTION: PaulaP to analyse RIFRAF discriminators cover CFA [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action08]

csma: We need the space of rule languages populated. We have 5 or 6 at the moment but need more
... Suggest a questionnaire to help populate the RIFRAF

+1

<sandro> +1

<AxelPolleres> I think I can take it over, if you formulate the concrete responsibility again.

<scribe> ACTION: Axel & sandro to prepare a questionnaire and organize it and keep track [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action09]

csma: The questionnaire allows people to complete an entry in the RIFRAF for a rule language oftheir choice
... If there are new criteria then the q may need to be modified and edited also
... the group thanks Axel

<AxelPolleres> sorry, still continued/

csma: action 20 continued
... action 22 continued
... we need a strawman semantics for RIFRAF

harold: We are working on this at the moment
... we have an update on semantics wrt positive conditions, e.g. model theoretic semantics
... now moving to horn logic and conditions

michael: it is possible to give a semantics inc conditions. parameterized by the language
... esp. model theoretic semantic based language

csma: can it be extended to negative conditions and built-ins?

michael: should be ok.
... am planning to enhance wiki

<csma> http://www.w3.org/2005/rules/wg/wiki/A.1_Basis%3A_Positive_Conditions

csma: Did pfps look at the positive conditions section above?

pfps: this seems to be a cut and paste of previous extensibility document

michael: parameterized by the model semantics of the language

pfps: My issue was that there was no contraint on mappings in and out of the RIFRAF

michael: to be done

pfps: Need requirements on mappings <-> RIFRAF

michael: yes

pfps: Need requirements on mappings <-> RIF
... needs to be moved from email discussion to the document

<Harold> pfps, can u go to A.1 Basis: Positive Conditions, click "Info", then "Diff", you'll see the changes afer the copy and paste.

<scribe> ACTION: michael to update semantics for positive conditions taking account of email discussion on contraints on RIF mappings [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action10]

csma: Mapping between existing rule languages and existing semantics

pfps: There are big gaps in the mapping at the moment.
... possible to have degenerate compliance

michael: Issue is pfps wants a unified luaguage for a super-language which is inherently small (without PRs, LP, etc)

pfps: I am not asking for that (I think)

<scribe> ACTION: pfps will describe loophole he sees in the semantics [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action11]

harold: Next step is to proceed to horn clauses
... We might be able to make progress in 2 weeks
... look at existing toc to see current approach

csma: better to lead the discussion by email

<scribe> ACTION: harold, michael to email ideas about extending semantics to cover horn logic [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action12]

csma: We need more volunteers to see how existing semantics and syntactic mappings are useful

alex There are many rule languages that belong to the PR categories. Existing semantic analysis not helpful to those languages

<PaulaP> I can also try to find time to map the cond lang to Xcerpt/XChange

<scribe> ACTION: alex will look at condition part of a sample PR language to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action13]

<scribe> ACTION: paulaP will look at condition part Xcerpt/XChange see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action14]

<LeoraMorgenstern> I will

<LeoraMorgenstern> Which language?

<LeoraMorgenstern> any version of prolog?

<LeoraMorgenstern> No, I have no preference

<LeoraMorgenstern> Okay

<scribe> ACTION: Leora will look at condition part of prolog to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action15]

<scribe> ACTION: Frank will look at condition part of Go! to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action16]

<AlexKozlenkov> go!

<LeoraMorgenstern> Okay, so you don't need me to do a version of Prolog, then.

The issue wrt prolog is the meta language

<PaulaP> +1

csma: Propose to adjourn

<AlexKozlenkov> OK, bye folks

Summary of Action Items

[NEW] ACTION: alex will look at condition part of a sample PR language to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action13]
[NEW] ACTION: AlexK will contact JSR94 community [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action05]
[NEW] ACTION: AlexKozlenkov will contact commercial rules vendors to take a look at the UCR [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action04]
[NEW] ACTION: Axel & sandro to prepare a questionnaire and organize it and keep track [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action09]
[NEW] ACTION: csma put organization of issues on next agenda [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action07]
[NEW] ACTION: csma will let SWCG know that there is an ongoing discussion and they will be updated [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action02]
[NEW] ACTION: Frank will look at condition part of Go! to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action16]
[NEW] ACTION: harold, michael to email ideas about extending semantics to cover horn logic [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action12]
[NEW] ACTION: Leora will look at condition part of prolog to see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action15]
[NEW] ACTION: michael to update semantics for positive conditions taking account of email discussion on contraints on RIF mappings [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action10]
[NEW] ACTION: PaulaP to analyse RIFRAF discriminators cover CFA [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action08]
[NEW] ACTION: paulaP will look at condition part Xcerpt/XChange see if semantic mapping is feasable, easy etc. [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action14]
[NEW] ACTION: pfps will describe loophole he sees in the semantics [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action11]
[NEW] ACTION: Sandro rename use and requirements into one category [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action06]
[NEW] ACTION: sandro: to fix the styles of the UCR [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action03]
 
[DONE] ACTION: 28 to [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2006/07/11 16:26:06 $