RIF Telecon 3 Oct 06 -- 3 Oct 2006

<ChrisW> agenda order is 1,7,2-6

<ChrisW> save agenda world-access

<sandro> rifbot, testing

<sandro> rifbot, help

<rifbot> See http://www.w3.org/2005/06/tracker/ for help (use the IRC bot link)

<csma> Chris, are you [IBM]?

<ChrisW> yes, i took care of it

<LeoraMorgenstern> Is anyone talking on the phone? I can hear nothing

<ChrisW> scribenick: Harold

<ChrisW> Scribe: Harold Boley

<ChrisW> http://lists.w3.org/Archives/Public/public-rif-wg/2006Sep/att-0081/19-rif-minutes.html

<csma> +1

<LeoraMorgenstern> yes

<LeoraMorgenstern> yes, I object

Minutes Sept 19: Leora found that not all names (of those who were Present) are mentioned, including her own one.

Christian: Names are at the end of those Minutes.

<ChrisW> MoZ are you there???

<csma> ACTION: leora to send email to MoZ to correct 19/9 minutes re names [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action01]

<rifbot> Created ACTION-136 - Send email to MoZ to correct 19/9 minutes re names [on Leora Morgenstern - due 2006-10-10].

<ChrisW> http://lists.w3.org/Archives/Public/public-rif-wg/2006Sep/att-0087/26-Sept-rif-minutes-c.html

<MoZ> ChrisW, yes I am

<MoZ> Sorry, I'm on another W3C telcon

Minutes Sept 26: Minor edits from Mala done, but not re-posted.

<ChrisW> Leora will send you a message, we need some edits to the ateendee list for the 9/19 telecon

<csma> ACTION: mala to repost edited minutes for 9/26 [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action02]

<rifbot> Created ACTION-137 - Repost edited minutes for 9/26 [on Mala Mehrotra - due 2006-10-10].

Admin

<PAULv> PaulV suggests review action 12 to Allen per minutes on last weeks minutes

F2F

<AlexKozlenkov> Chris I nee that visa letter ASAP

F2F4: Register everyone!

Although no real deadline, register asap.

Hotels within walking distance.

F2F5

<PAULv> +1 Antartica

<hak> +1

<pfps> +1

<MalaMehrotra> +1

<DavidHirtle> +1

Chris: During winter break at MIT, there will be a mini-conference, with which we could combine our F2F5 in Cambridge.

<LeoraMorgenstern> I'm afraid the minutes will reflect that as +1's for Antarctica, not for Banff!

<sandro> csma: Banff is likely to cause some conflict with W3C AC meeting.

F2F6: Sandro: Banff (WWW conf), Innsbruck (ESWC), or 3rd place?

Axel: ESWC would be doable.

Sandro: W3C mini-conference may be difficult, room-wise.

<LeoraMorgenstern> Chris, perhaps IBM could host it?

DavidR: Bristol is remote possibility as well.

<AxelPolleres> As mentioned, we think about it co-located with ESWC, but we need to check back first.

Chris: IBM is remote possibility as well.

<PAULv> Qu: what numbers of people would be expected at F2F5/6?

Chris: 30-40 people.

Sandro: People should face each other (not: classroom-style).

Harold: 'U-shape'.

<csma> action Sandro to send an email message calling for F2F5 and F2F6 hosting proposals

<csma> ACTION: sandro to send email message calling for F2F5 and F2F6 hosting proposals [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action03]

<rifbot> Created ACTION-138 - Send email message calling for F2F5 and F2F6 hosting proposals [on Sandro Hawke - due 2006-10-10].

Liaison

No new activity.

Use Cases & Requirements

<scribe> ACTION: 41 to Explain blackbox: Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action04]

<scribe> ACTION: 72 to Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action05]

<csma> ACTION: 38, 41, 72, 87 continued [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action06]

<rifbot> Sorry, couldn't find user - 38,

<scribe> ACTION: 87 to Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action07]

http://www.w3.org/2005/rules/wg/wiki/CORE

<scribe> (new page)

<scribe> ACTION: Done. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action08]

<rifbot> Sorry, bad ACTION syntax

<ChrisW> action 102 is done

<csma> action 105 continued

<csma> action 106 continued

Action 105/106: Continued.

<scribe> ACTION: 113 to Completed. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action09]

<scribe> ACTION: 114 to Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action10]

Action on Christian: Co-location. Continued.

<scribe> ACTION: 125 to Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action11]

Action 126 (on Allen): sent email with possible text.

<scribe> (Completed)

Action 127 (on Leora): linked, explanations, additions (in comments), without rules.

<scribe> Completed.

<scribe> ACTION: 128 to Done. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action12]

Change action on Francois to one on Paula.

Action 129 (on Allen): Continued.

<AxelPolleres> continued

Action 130 (on Axel): Continued

Action 131-133: Continued.

<scribe> ACTION: 134 to Continued. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action13]

<scribe> ACTION: 135 to Completed. [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action14]

<sandro> Yay Harold! Yay Leora!!

UC->Req links for 1, 6, 9 are ready.

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

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2006Oct/0003.html

Intro text on Allen: Updated wiki page (mentioned in email).

Says, there are some reqs from the charter (XML etc.).

Tried to capture comments of Christian.

<ChrisW> http://www.w3.org/2005/rules/wg/wiki/UCR/Introduction

<DaveReynolds> Looks good to me

Chris: New paragraph: Looks good.

Leora, Chris: Would be good to say which reqs are motivated by UCs, functional, chartered.

Allen: Wait a bit until we've more data.

Christian: Regarding functional reqs, should wait until finished analysis of UCs.

Chris: Perhaps 'reverse engineer' which ones are functional, which ones UC-motivated.

<sandro> [ Allen, I have one problem with the new paragraph which is the use of the definite article before "RIF". Does your browser display web pages written in the HTML or the XHTML, or both? Did you ever program in the LISP? .... :-) ]

<ChrisW> I'll still be watching the IRC out of the corner of my eye

Christian: For my UC, Dave's comment was taken into account.

<LeoraMorgenstern> http://www.w3.org/2005/rules/wg/wiki/UCR/Ruleset_Integration_for_Medical_Decision_Support

<Allen> [Sandro, I will change it]

UC6: Leora: Put assumptions in the links. Ex: Embedded metadata.

<Allen> [Sandro: Actually the usage "The RIF" occurs all over the place. Do you want to get rid of that everywhere?]

<AxelPolleres> sorry, my headset was mutet... strange

Sandro: Part of Leora's text is meant for discussion, perhaps put it in italics.

Christian: Ex: the first UC: You should clarify the req from the point of view of this UC.

Leora: It's vague what "Different intended semantics" expresses.

Christian: All others who do UC->Req linkages should propose new wordings for a Req if original Req is unclear.

Leora: Could be difficult if everybody does their own interpretation.

Sandro: Send email to the list proposing new Req wordings.

Christian: Problem with the 'XML data' Req: Unclear what it really means.

Leora: My understanding of OWL compatibility: If you look at a KB in OWL, there should be a way of translating it to RIF.

Christian: Maybe OWL KB can be shared in another way?

Leora: Because you don't want to lose 'OWLness'?

Christian: OWL fragments as black boxes?
... How to cover any kind of data.

Sandro: Translate to RIF.

Christian: In most general sense: RIF would provide alternate serialization of OWL and RDF?

Sandro: Yes, from charter.

Allen: Besides clarifying what a Req is, UC may lead to a subcase of a general Req: Another reason why clarification is important.

<FrankMcCabe> +q

DaveR: RDF data:
... Did post what it means. Gist: If rule language carries RDF-compatible data, there need to be a mapping.

Christian: Agree for XML types.
... Maybe replicate for RDF and OWL types.
... Difficult, however, for data in (whatever) format, because only for XML there is XML Schema.

<sandro> FrankMcCabe, can you mute?

Dave: You still need to explain what correspondence is.

Christian: Doesn't imply that RIF is serialization for RDF or OWL. Instead, give user a way to specify a data format (e.g. Integer), without necessarily having a translator.

Sandro: RIF can convey rule aspects of other languages, e.g. RDF triples as facts.

Christian: Will take action on clarifying certain Reqs. Maybe others can then try to explain them better.

<DaveReynolds> Christian - perhaps you respond to my suggested clarification already posted?

Christian: From standardization point of view it's not a good idea to capture something that can already be done with another spec.

<csma> +1 to Dave

Leora: Coverage not so clear: RIFRAF often uses either "... or ..."
... Other issue: how does whole ruleset look like? Eg. disjunction allowed? Vague descriptions.

<DaveReynolds> +1 to Leora, these high level UCs should not motivate individual coverage requirements

<PAULv> PAULv - apologies - must drop off call now

Allen: Not prohibited to use rules in English.

UC 9: Harold:

http://www.w3.org/2005/rules/wg/wiki/UCR/BPEL_Orchestration_of_Rule-Based_Web_Services

Harold: In Coverage picked one alternative for relevant RIFRAF discriminators such as "Syn 5: Slotted Arguments, since ..."

Christian: Should we look more carefully at the rules, and the need for their interchange?

Gary: Yes, exact need/reason for interchange is a weak point of UC 9, and also of some other UCs.

Allen: UC 9 Reqs look OK. But some further Reqs may be missing, which we would then need to add ourselves or send to Harold or Gary.

Christian: We now need to move this to email discussion.

RIFRAF

Christian: Axel is editor, but we have not decided yet on a possible ontology (OWL) language for RIFRAF.

Axel: Perhaps we could also continue to edit the current document.
... But others should jump in and help.

Frank: Did work on type system.

<sandro> pointer, Frank?

<LeoraMorgenstern> So would I

Christian: Can we have a task force to do an ontology? Who will help?

<sandro> task force for OWL version of RIF RAF: Sandro, Frank, Allen, ....

<sandro> and Axel, but not so much in the next couple weeks.

<sandro> and Leora

<FrankMcCabe> Pointer to RIFRAF questionnaire on the WIKI:

<FrankMcCabe> http://www.w3.org/2005/rules/wg/wiki/Rulesystem_Arrangement_Framework

<FrankMcCabe> See section 6

Christian: Can Axel do edit?

Axel: Not until Oct 20.

<sandro> ACTION: Leora to start organizing RIFRAF Ontology Task Force (using public-rif-wg mailing list) [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action15]

<rifbot> Created ACTION-139 - Start organizing RIFRAF Ontology Task Force (using public-rif-wg mailing list) [on Leora Morgenstern - due 2006-10-10].

<LeoraMorgenstern> Axel, okay, that's fine.

<LeoraMorgenstern> yes, I'll keep you in the loop

Technical Design

Christian: We have to move to XML syntax.

Christian: DanC's email raised a number of questions. Other implementations will surely raise more questions: Extensibility, types, etc.
... Large part of upcoming F2F4.

<sandro> ACTION: Harold to construct a trial RIF implementation in J-Drew [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action16]

<rifbot> Created ACTION-140 - Construct a trial RIF implementation in J-Drew [on Harold Boley - due 2006-10-10].

<hak> I can show what I have done with my Jacc-bases XML annotation for RIF generation

Christian: Should be made iteratively, because of variations in syntax.

Hassan: Will release implementation once internally approved.

<sandro> ACTION: Hassan to release Jacc (current version, while he keeps working on his improvements) [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action17]

<rifbot> Created ACTION-141 - Release Jacc (current version, while he keeps working on his improvements) [on Hassan Ait-Kaci - due 2006-10-10].

<sandro> ACTION: Hassan to produce an IRL implementation of RIF [recorded in http://www.w3.org/2006/10/03-rif-minutes.html#action18]

<rifbot> Created ACTION-142 - Produce an IRL implementation of RIF [on Hassan Ait-Kaci - due 2006-10-10].

Sandro: While DanC cannot come to F2F4, Sandro will be able to present the translator from N3-rules syntax to the RIF XML syntax for Horn rules.

<sandro> +1 adjourn

- DRAFT -

RIF Telecon 3 Oct 06

3 Oct 2006

Attendees

Contents