W3C

- DRAFT -

RIF telecon 21 July 2009

21 Jul 2009

Agenda

See also: IRC log

Attendees

Present
AdrianP, DaveReynolds, Gary Gary_Hallmark, Harold, Sandro, Stella_Mitchell, csma
Regrets
Michael_Kifer, Leora_Morgenstern, Axel_Polleres, Chris_Welty
Chair
Christian de Sainte Marie
Scribe
Adrian Paschke

Contents


<csma> agendum admin

<csma> Hello CorentinF! Who are you?

<CorentinF> hi

<csma> Hi

<csma> Who are you?

<CorentinF> i'm a trainee at inria, i work on a rif-bld implementation

<csma> Are you a member of the WG?

<CorentinF> no

<CorentinF> should i leave?

<csma> Well, telecons are for WG members only. But if you can tell me a little bit more about your implementation, we will have that on the record...

<csma> And, then, you can leave :-)

<csma> What is your time frame?

<csma> For implementation, I mean

<csma> Corentin, what kind of implementation are developing?

<csma> Translating BLD to/from what language?

<csma> What is your target date for a working prototype?

<csma> scribenick: AdrianP

<CorentinF> well for now it's a very experimental work, meant to get an interperter for presentation syntax

<csma> You don't use the XML syntax?

<AdrianP>CorentinF you might be interested in this demo of the RuleML-2008 Rules Challenge by IBM implementing RIF BLD

http://ruleml-challenge.cs.nccu.edu.tw/?q=node/22

<AdrianP>At RuleML-2008 there were also further demos and presentations about RIF

<CorentinF> we planned to make an xml concrete parser as well, but as it's to be integrated with a research project software, the "interchange" aspect is not really critical

<csma> Great (although only the XML syntax is normative, of course)

<csma> Corentin, can you sent a comment to the public comment list, so that we have your email address. And any comments you have about BLD would be great, too

<Harold> http://lists.w3.org/Archives/Public/public-owl-comments/

<csma> the public comment address is: public-rif-comments@w3.org

<CorentinF> sure

<csma> Corentin, you will have to leave, now.

<Harold> http://lists.w3.org/Archives/Public/public-rif-comments/

<csma> We will keep in contact via the mailing list

scribenick, AdrianP

<csma> next item

Admin

<csma> PROPOSED: accept minutes of telecon July 7

<StellaMitchell> http://lists.w3.org/Archives/Public/public-rif-wg/2009Jul/att-0021/2009-07-07-rif-minutes.html

<csma> RESOLVED: accept minutes of telecon July 7

resolved minutes accepted

csma: next telecon to review the comments?

<Gary> I will be unable to attend on Aug 4

I could attend first week of August

Harold: Chris mentioned that we might need a reasonable number of comments

I will be travelling next week, but will try to call in

csma: next telecon 28th of July

<csma> Sandro, will you join the telecon?

<csma> next item

<sandro> Sorry, csma, no.

<sandro> (I hope you saw my e-mail about this.)

Liaisons

csma: liaisons

<csma> Sandro, is there any condition re amount or quality of public comments for moving to CR?

<sandro> There is no defined line. We just have to make the case that there's been wide public review.

<csma> Sandro, as OWL liaison, could you push them to comment?

<sandro> It MAY be enough to have gotten the word out and be ignored. It would be much better to have simple messages along the lines of "this looks okay."

<sandro> ACTION: sandro to get an OWL-WG comment on RIF [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action01]

<trackbot> Created ACTION-892 - Get an OWL-WG comment on RIF [on Sandro Hawke - due 2009-07-28].

Adrian: W3c HCLS BIORDF will take a look for a micro array experiemts use case

<csma> next item

Action review

action 889 continued

<trackbot> Sorry, couldn't find user - 889

action 885 closed

<trackbot> Sorry, couldn't find user - 885

action-885 closed

<trackbot> ACTION-885 Answer FAQ 3.2-3.3 and 3.5 for RL closed

action-884 closed

<trackbot> ACTION-884 Answer FAQ 3.6 closed

action-881 closed

<trackbot> ACTION-881 Contact Aberdeen colleagues. closed

action-876 closed

<trackbot> ACTION-876 Contact Said Tabet. closed

action-874 closed

<trackbot> ACTION-874 Contact Guido Governatori. closed

action-872 closed

<trackbot> ACTION-872 Contact Igor Mozetic. closed

action-869 closed

<trackbot> ACTION-869 Contact Giorgos closed

action-868 closed

<trackbot> ACTION-868 Contact Michael Sintek closed

action-867 closed

<trackbot> ACTION-867 Contact Luis Polo. closed

action-866 closed

<trackbot> ACTION-866 Contact Markas? closed

action-861 closed

<trackbot> ACTION-861 Announce inside oracle, drules, jess closed

action-860 closed

<trackbot> ACTION-860 Announce to hp internal and jena-dev closed

action-856 closed

<trackbot> ACTION-856 Announce to RuleML, NRCC, UNB closed

action-853 closed

<trackbot> ACTION-853 Update XML syntax for Core and BLD to use xml:lang attribute for consts closed

action-852 closed

<trackbot> ACTION-852 Review Multiple* test cases closed

action-852 pending review

action-845 complete

action-845 pending review

action-842 closed

<trackbot> ACTION-842 Provide some feedback on new safeness by tomorrow closed

action-784 closed

<trackbot> ACTION-784 Talk to Mark Proctor about RIF implementation closed

action-782 closed

<trackbot> ACTION-782 Talk to ontoprise about RIF implementation & timeframe closed

action-780 closed

<trackbot> ACTION-780 Refactor BLD schema to import Core (help from cke) closed

<csma> next item

Last call publications

<Harold> We had two comments from David Mott:

<Harold> http://lists.w3.org/Archives/Public/public-rif-comments/2009Jun/0000.html

<Harold> http://lists.w3.org/Archives/Public/public-rif-comments/2009Jun/0001.html

Harold: two comments by IBM
... we should have a Wiki page where we collect all public comments

<Harold> Another comment from Nick Bassiliades, Greece, is pending: he could not yet send it for list-technical reasons.

csma: wheneve we here about RIF implementations we should ask people to send a comment
... comment like I read the spec, is better than no comment
... encourage all contacts to send comments
... Implementations?
... should start listing implementation plans

Harold: Alexander Riazanov plans to implement BLD in his Vampire Theorem Proofer

<csma> http://www.w3.org/2005/rules/wiki/Implementations

<DaveReynolds> http://www.w3.org/2005/rules/wg/wiki/Implementation_plans

<DaveReynolds> That one isn't empty

action, csma change implementation page

<Harold> The current RIF Implementation plans table is outdated.

<scribe> ACTION: csma to change implementation page [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action02]

<trackbot> Created ACTION-893 - Change implementation page [on Christian de Sainte Marie - due 2009-07-28].

<Harold> I came up with this:

<Harold> Possible RIF Implementations

<Harold> (probably not TextGarden from above Implementation_plans):

<Harold> dlvhex: Core (see above Implementation_plans)

<Harold> Flora-2: BLD (Michael Kifer et al.)

<Harold> Oracle: PRD (Gary Hallmark et al.)

<Harold> IBM/ILOG: PRD (CSMA et al.)

<Harold> W3C: Core (translator to N3, Sandro Hawke)

<Harold> RuleML Inc: BLD (translator to Sigma Kee Vampire, Alexandre Riazanov)

<Harold> FU Berlin: PRD (translator to Prova / Reaction RuleML, Adrian Paschke et al.)

<Harold> NRC: Core (translator to OO jDREW / Datalog RuleML, Harold Boley et al.)

csma: can we put it on the Wiki page

<scribe> ACTION: csma add possible/potential implemenations on the Wiki page (labeled as potential implementations) [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action03]

<trackbot> Created ACTION-894 - Add possible/potential implemenations on the Wiki page (labeled as potential implementations) [on Christian de Sainte Marie - due 2009-07-28].

<Harold> I don't know if possible Ontoprise and Flora-2 implementations might be related.

csma: did not get feedback from Ontoprise yet

<Harold> Mark Proctor's DROOLS work should be highly relevant for PRD.

Dave: Pascal and Markus are currently not able to do RIF implementations due to lack of resources

<csma> CTIC interested in trying a PRD implementation for SPARQL + construct

<csma> CTIC also interested in developed a RIF editor

<csma> Adrian possibly developing an implementation for Tibco

Adrian: translation to Prova
... academic collaboration project with TIBCO, working on a translator

Gary: work on a PRD implementation

csma: heared that ontoprise and Vienna are working on a new dialect using FLD

Harold: non-monotonic dialect or just using FLD?

csma: a new dialect using FLD
... CoronteF at INRIA working on a BLD implementation

Harold: wonder if Stella uses one of the tools from Hasan or Sandro

Stella: yes, I use the tool from Hasan

Harold: can everyone download it

Stella: can look-up the link

<Harold> Hassan could be asked if we can publicly point to his tool

Harold: meant that it can count a first implementation

<Harold> as a first implementation.

<Harold> (if only the syntax part, but it already helped Stella et. al a lot.)

csma: you mentioned an IBM demo

Adrian: yes, that was demo given last year at the Rules Challenge at RuleML-2008
... need to see if it can count as a BLD implementation

I will ask her if she will also send comments about RIF

<csma> next item

<scribe> ACTION: AdrianP ask participants of RuleML-2008 which did RIF demos to send comments [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action04]

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

<scribe> ACTION: Adrian ask participants of RuleML-2008 which did RIF demos to send comments [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action05]

<trackbot> Created ACTION-895 - Ask participants of RuleML-2008 which did RIF demos to send comments [on Adrian Paschke - due 2009-07-28].

<csma> http://www.w3.org/2005/rules/wiki/RIF_FAQ#How_do_I_serialize_RIF_rules_in_RDF.3F

csma: how do we proceed for FAQ; just approve them?

Harold: it is an inoffical page
... so enough to just approve them
... under 3.5. linked "defines" is hard to spot, should be made more explicit

csma: there is a commercial implementation by Ontoprise
... Ontobroker supports RIF
... should publish FAQ as early as possible
... maybe should take it as is and publish even without all quesions answered

Harold: could everyone complete the FAQ questions

Stella: 4.4 wasn't sure about it

csma: Dave suggestion to review it later is good
... 4.4. could be "not yet"

<DaveReynolds> I've made the link to SWC more explicit in FAQ 3.5

<csma> next item

<sandro> the FAQ is already published.

<csma> Sandro, it is not referenced from the WG front page

<sandro> yes it is.

<DaveReynolds> Yes it is right at the top

<Harold> FAQ is published but it says: "This is an unofficial draft FAQ"

csma: shall we ask someone from the working group to review it

Sandro: sounds good

RIF + XML data

csma: added editors notes
... looked at Xpath 2.0 - reworked document relates to more general elements in the XML stack

Sandro: we have DTB built-ins, we could have standard built-ins for XML, i.e. Xpath built-ins

csma: should be based on the XML data model

<csma> next item

Test Cases

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

Sandro: maformedLists

Stella: could nothing to be concluded from the test cases?
... csma is saying wording is similar to be similar to ArbitraryEntailment
... ArbitraryEntailment test case

Sandro: malformed lists are structures too

<sandro> "Malformed lists are structures, too. There's nothing illogical about them."

<Harold> A few months ago, "malformed lists" were renamed into "general open lists".

Sandro: is a list with a tail which is not a list

<Harold> http://www.w3.org/2005/rules/wiki/BLD#Terms

Harold: part 4

<Harold> List terms.

<StellaMitchell> s/could nothing be conclused from the test cases/is there anything that is entailed by the premise in this test case/

<Harold> Such general open lists correspond to Lisp's dotted lists [Steele90].

<StellaMitchell> s/could nothing be concluded from the test cases/is there anything that is entailed by the premise in this test case/

Stella: maybe a longer description helps to make it more clear

csma: if there is not a positive entailment it means there is a contradiction
... should at least entail the premise

<sandro> description: "The tail of a list in BLD doesn't have to be a list. It can be like a LISP dotted pair; it's just another kind of structure."

Gary: think the premise is entailed

<sandro> if List ( ?x | ?y ) and is-string(?y) then is-funky-structure(?x)

Harold: problem is when there is a variable we cannot control if it is bound to a list or another term

<sandro> csma: if List ( ?x | ?y ) and is-list ?y then is-list List( ?x | ?y )

csma: shall we change description of test case?
... test says it is not a syntax error if the tail is not a lists

Sandro: premise is in fact a list, according to the definition of BLD

<sandro> although as I recall from LISP it's called a "dotted pair".

<Harold> Lisp terminology: dotted pair: (a . b) -- dotted list: (a b c . d)

<Harold> However, (a b c . d) is just 'presentation syntax' for nested dotted pairs (a . (b . (c . d)))

<csma> PROPOSED: extend 10 mn to conclude on MalformedLists

<DaveReynolds> Ok to extend

+1

<csma> RESOLVED: extend 10 mn to conclude on MalformedLists

<Harold> "Note that the last condition above restricts Itail only when its last argument is in Dlist. If the last argument of Itail is not in Dlist, then the list is a general open one and there are no restrictions on the value of Itail except that it must be in Dind. "

<Harold> http://www.w3.org/2005/rules/wiki/BLD#Semantic_Structures

<Harold> 5. Ilist and Itail are used to interpret lists

Sandro: premise does not entail itself
... need to ask Jos

<sandro> ACTION: sandro to ask Jos about http://www.w3.org/2005/rules/wiki/MalformedLists [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action06]

<trackbot> Created ACTION-896 - Ask Jos about http://www.w3.org/2005/rules/wiki/MalformedLists [on Sandro Hawke - due 2009-07-28].

AOB

Adrian: plan to have a W3C RIF workshop at RuleML-2009 (http://2009.ruleml.org) with a tutorial about RIF PRD by Christian and demonstration of RIF implementations.

Adrian: and Sandro will give a keynote about RIF at RuleML-2009

Adrian: intention is to reach out to the business rules community and spread the word about the new W3C RIF standard

Adrian: will send further info to the working group

Summary of Action Items

[NEW] ACTION: Adrian ask participants of RuleML-2008 which did RIF demos to send comments [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action05]
[NEW] ACTION: AdrianP ask participants of RuleML-2008 which did RIF demos to send comments [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action04]
[NEW] ACTION: csma add possible/potential implemenations on the Wiki page (labeled as potential implementations) [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action03]
[NEW] ACTION: csma to change implementation page [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action02]
[NEW] ACTION: sandro to ask Jos about http://www.w3.org/2005/rules/wiki/MalformedLists [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action06]
[NEW] ACTION: sandro to get an OWL-WG comment on RIF [recorded in http://www.w3.org/2009/07/21-rif-minutes.html#action01]
 
[End of minutes]