W3C

RIF Telecon 18-Nov-08

18 Nov 2008

Agenda

See also: IRC log

Attendees

Present
Hassan Ait-Kaci, Jos de Bruijn, Harold Boley, Gary Hallmark, Sandro Hawke, Michael Kifer, Stella Mitchell, Leora Morgenstern, Adrian Paschke, Axel Polleres, Dave Reynolds, Chris Welty
Regrets
Paul Vincent
Chair
Christian de Sainte-Marie
Scribe
Leora Morgenstern

Contents


 

 

Admin

<csma> item 1

<csma> PROPOSED: accept minutes of telecon November 11

<sandro> scribenick: LeoraMorgenstern

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2008Nov/att-0042/2008-11-11-rif-minutes.html

<csma> RESOLVED: accept minutes of telecon November 11

csma: any agenda amendments?

Harold: Dave Reynolds and I would like to bring in two topics from Core discussion yesterday to agenda.
... with respect to pattern matching vs. unification and also of # and ##

<sandro> action-1?

<trackbot> ACTION-1 -- Leora Morgenstern to set up draft proposal on what we mean by FOL -- due 2006-04-30 -- CLOSED

<trackbot> http://www.w3.org/2005/rules/wg/track/actions/1

<sandro> (just testing trackbot)

csma: Discussing these would take more than 5 minutes. More urgent than DTB and UCR?

chris: Probably not, so will only discuss if there is time.

Harold: okay

Liason

csma: Any news, comments from liaisons?

csma: was RDF text published?

sandro: no, due to publishing snafu; still debating, also, whether to send it out now, or later.

csma: anything new wrt XBRL?

sandro: <get answer wrt XBRL>

Public Comments

3 comments need to be responded to. There are draft responses or beginnings of responses.

chris: I have to deal with 2 public comments; Christian is supposed to deal with RAK (Kowalski).

csma: I will address this once I have gone through the paper http://www.doc.ic.ac.uk/~rak/papers/conditionals.pdf; will ask Gary for support.

Harold: text in BLD has changed somewhat [in accord with O'Keefe's comment that KIF's influence ought to be mentioned].

chris: I'm the one who is supposed to draft the response.

Action Review

Action 655 on Dave to update EBNF to match BLD is done.

<trackbot> Sorry, couldn't find user - 655

654: done

action 653 to open issue on polyadic functions per note in DTB 3.5 is continued

<trackbot> Sorry, couldn't find user - 653

652 to write some more pathological test cases for potentially confusing uses of iri-string predicate --- continued

<Harold> Axel, Yes, we should have multi-arity built-ins (e.g. max and min).

651: done

650: continued (see above)

648 & 647 on Axel ( to add editor's note on xml schema compatibility, and add editor's note regarding status of finite/infinite characters) turned out to be the same action, and is/are done.

646 on Gary to draft a response to RAK: continued

645: Chris to send WL response is continued

644: Draft comment to OK: continued.

643: done

<AxelPolleres> harold, a) we have no resolution on this, b) this also aplies to e.g. fn:concat, not to new, not yet agreed built-ins

<AxelPolleres> do we have a resolution on min/max ?

637 on sandro to generate test case manifests: basically done, although Stella and Sandro still have to work out some details.

633 on Chris to discuss to how specify prefixes on email: continued, needs to be moved to PS task force.

604 on Axel: continued for now.

592: continued

588 on Leora: continued; part of bigger effort to collect UCR cases that can't be represented currently in RIF.

573, 579 on sandro: continued

565 on sandro (access for Adrian to CVS) is done.

564 on Hassan: needs to be moved to PS task force, continued

439, 152 continued.

Pending reviews: postponed to agenda discussion

Task Force Updates: Core, PS, PRD

Task force updates.

csma: PRD task force: we are still discussing our next working draft; discusson will probably be concluded later today.

chris: PS task force: We met on Friday. We split issues into 2 categories: (1) making the syntax more presentable (2) making the syntax less ambiguous. We came up with a set of changes that should reduce the ambiguities. Everyone including Hassan was content. Hassan is working through some implementational issues before changes can be tested.
... still a lot of disagreement about what's presentable. Hard to judge presentability, especially with a group so used to doing things in particular ways.
... easier to reduce ambiguity, because there are clear technical requirements.

<StellaMitchell> I will send to whole group

<ChrisW> done

<StellaMitchell> ok

harold: perhaps we should look at OWL working group since many in RIF group will want to use RIF rules with OWL ontology.

<Hassan> +1 on Harold's suggestion to get inspired on OWL's abstract syntax

<Harold> http://lists.w3.org/Archives/Public/public-rif-wg/2008Nov/0103.html

Harold: Core task force. Minutes linked above.
... Discussed member and subclass. Consensus: have # in rule bodies.
... More constructs on Core defined by referring to BLD. Process not finished, but should be done by Monday.

csma: Does this mean implementing Core will require reading and understanding BLD?

harold: yes, and PRD too.

csma: Doesn't that make things more difficult for people who just want to implement Core?

harold: yes, but necessary to facilitate maintenance.

dave: yes, otherwise there would be tremendous amounts of duplication.

csma: why not explicitly specify in Core, and have BLD and PRD make reference to it?

harold: Because BLD is further advanced, and last call, while Core is not yet stable.

<Harold> PRD colleagues, could you pleas briefly bring up if http://www.w3.org/2005/rules/wiki/Core#Intersection_of_RIF-PRD_and_RIF-BLD is fine for the FPWD of Core?

harold: and Core is less than the intersection of PRD and BLD.
... easier to extend BLD and PRD to Core.

<Harold> Yes, less than intersection.

UCR

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2008Nov/0041.html

adrian: I added two new requirements to UCR
... First: inter-dialect and intra-dialect interoperability
... this had never been explicitly stated

<csma> RIF must allow intra-dialect interoperation, i.e. interoperability between

<csma> semantically similar rule languages (via interchange of RIF rules) within

<csma> one dialect, and it should support inter-dialect interoperation, i.e.

<csma> interoperation between different dialects with maximum overlap.

csma: discussion? Any objections to this requirement?

Dave: Just some comments on how this got lost: We had this requirements structure, and interoperatibility was one of the high-level goals.
... (i.e., high-level goal rather than requirement)

csma: So perhaps interoperabilty should be moved to a high-level requirement. Structure things so that there's no overlap.

<csma> RIF must support internationalized text - that is, text that additionally

<csma> conveys information in terms of a language tag.

adrian: Second requirement: internationalized text (see above).

csma: any discussion or objection to internationalized text requirement?
... we'll have a resolution next week about this.

sandro: Is interoperability just restricted to common substrates of the dialects?

<StellaMitchell> "maximum overlap" had been my comment

csma: summary: resolution next week regarding internationalized text; while interoperatibility still under discussion

Publication Plans: Core, PRD, Test Cases, DTB

csma: according to plan UCR. Core, Test, PRD, DTB due to be frozen today.

chris: UCR has been reviewed

adrian: some open examples and work needs to be done on the examples
... but otherwise ready for next publication

harold: There will be some changes before we can freeze it. Our target for freezable core is Monday, Nov. 24

<Michael_Kifer> next Mon should be ok

harold: gary, michael, and dave may need more time

<Harold> http://www.w3.org/2005/rules/wiki/Core

<Michael_Kifer> next Mon should be ok

<LeoraMorgenstern> I can review Core

<DaveReynolds> The EBNF updates are done modulo decisions on #/## and external functions

<ChrisW> ACTION: csma to review Core by two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action01]

<trackbot> Created ACTION-656 - Review Core by two weeks [on Christian de Sainte Marie - due 2008-11-25].

<ChrisW> ACTION: leora to review core by two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action02]

<trackbot> Created ACTION-657 - Review core by two weeks [on Leora Morgenstern - due 2008-11-25].

csma: PRD: we still have the question of # ##, but not holding up PRD since it's a working draft.
... xml, external data models, also not in this draft.
... should be able to have this WD done by Friday.
... so we'll freeze by Friday.

<ChrisW> hassan, would you have time to review PRD?

<ChrisW> ACTION: mkifer to review Core in two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action03]

<trackbot> Created ACTION-658 - Review Core in two weeks [on Michael Kifer - due 2008-11-25].

<ChrisW> ACTION: hassan to review PRD - due in two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action04]

<trackbot> Created ACTION-659 - review PRD [on Hassan Ait-Kaci - due 2008-11-18].

<Hassan> 3^D

<sandro> action-659 due 2008-12-03

<trackbot> ACTION-659 review PRD due date now 2008-12-03

stella: Test is ready to freeze. Still needs more work, especially wrt format, but Sandro feels it's okay for a first pass.

<ChrisW> ACTION: sandro to review test - due Dec 02 [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action05]

<trackbot> Created ACTION-660 - review test [on Sandro Hawke - due 2008-12-02].

<ChrisW> ACTION: gary to review test - due Dec 02 [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action06]

<trackbot> Created ACTION-661 - review test [on Gary Hallmark - due 2008-12-02].

<LeoraMorgenstern> Stella, Adrian, should there perhaps be a pointer in the Test doc to the spreadsheet that I prepared regarding the test cases?

DTB

axel: DTB: Edtiing editor's note still needs to be done. Action 604 needs to be done.

chris: DTB has been reviewed.

axel: making changes in response to reviews. Should be done by the end of the week.
... everything not done is marked with editor's notes. I will leave editor's notes, but may still make some changes to improve situation.

jos: Not all my comments resulted in editor's note, e.g., mapping , data type conversions.

axel: I will add editor's notes as needed.

DTB

<csma> http://www.w3.org/2005/rules/wg/track/issues/67

This has already been done in above discussion.

Next topic to discuss: Issue 67

Issue 67: inconsistency in the way numeric-less-than operator is used.

josb: not natural to compare strings in the same way as numbers: numbers have a natural ordering

<Harold> Strings have lexicographic ordering.

<AxelPolleres> s/Alex/Axel/

Gary: What's more important for strings is equal and non-equal, rather than ordering.

<AxelPolleres> from fn:compare (xpath-functions): This function, invoked with the first signature, backs up the "eq", "ne", "gt", "lt", "le" and "ge" operators on string values.

csma: Also, issues 79, 80, 81.

csma: propose to have equal and non-equal for all those types (strings, etc.) and to leave the greater than and less than issues for now.

Axel: I'd prefer to mark this with an editor's note, rather than remove the gt, lt, le, ge orderings in those cases.
... issue 67, and thus the editor's note, was only on strings.

<DaveReynolds> Sorry I have to leave slightly early: no objection to =/!= on strings; regarding issue 80 I don't think this is yet closable

<AxelPolleres> I see no point in closing "half" of the issue...

<AxelPolleres> ... to be honest.

<csma> PROPOSED: Add equal and not-equal builtins for string

Axel: why is equal and not-equal less contentious than gt and lt?

<Harold> I think for DTB, which is a kind of taxonomy of mostly independent features (built-ins) we should keep reasonable built-ins, so DTB has a chance to become the ground layer of built-ins for the (Semantic) Web.

<josb> no, as it stands DTB is a part of BLD

<AxelPolleres> As i typed earlier, fn:compare in Xquery/Xpath backs up eq, ne, gt, and lt... so if we only want to have half of those... fine.

<AxelPolleres> http://www.w3.org/TR/xpath-functions/#func-compare

<ChrisW> ACTION: csma to put string= proposed resolution on next week's agenda [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action07]

<trackbot> Created ACTION-662 - Put string= proposed resolution on next week's agenda [on Christian de Sainte Marie - due 2008-11-25].

<Hassan> +1 to adjourn

AOB

Adrian is scribe for next week

No other business.

Summary of Action Items

[NEW] ACTION: csma to put string= proposed resolution on next week's agenda [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action07]
[NEW] ACTION: csma to review Core by two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action01]
[NEW] ACTION: gary to review test - due Dec 02 [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action06]
[NEW] ACTION: hassan to review PRD - due in two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action04]
[NEW] ACTION: leora to review core by two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action02]
[NEW] ACTION: mkifer to review Core in two weeks [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action03]
[NEW] ACTION: sandro to review test - due Dec 02 [recorded in http://www.w3.org/2008/11/18-rif-minutes.html#action05]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/11/18 17:29:44 $