- From: Francois Bry <bry@ifi.lmu.de>
- Date: Fri, 27 Jan 2006 13:08:37 +0100
- To: W3C RIF WG <public-rif-wg@w3.org>
RIF Working Group
telecon of 24 Jan 2006
Minutes
Attendees (in alphabetic order)
Allen Ginsberg (Allen_Ginsberg)
Andreas Harth (Andreas_Harth)
Bill Andersen (BillAndersen)
Chris Welty (ChrisW)
Darko Anicic (Darko)
David Hirtle (David_Hirtle)
Deborah Nichols (Deborah_Nichols)
Deepali Khushraj (Deepali_Khushraj)
Donald Chapin (DonaldC)
Edward Barkmeyer (Ed_Barkmeyer)
Evan Wallace (Evan_Wallace)
Francois Bry (FrancoisBry)
Frank McCabe (FrankMcCabe)
Gary Hallmark (GaryHallmark)
Giorgos Stamou (GiorgosStamou)
Harold Boley (Harold)
Hassan Ait-Kaci (Hassan_Ait-Kaci)
Igor Mozetic (Igor_Mozetic)
Jeff Pan (JeffPan)
Jeremy Caroll (JeremyCarroll)
John Hall (John_Hall)
Jos De Roo (JosDeRoo)
JosDeBruijn (josb)
Leora Morgenstern (LeoraMorgenstern)
Markus Krötzsch (MarkusK)
Michael Kifer (Michael_Kifer)
Michael Sintek (Michael_Sintek)
Mike Dean (Mike_Dean)
Minsu Jang (MinsuJang)
Paula-Lavinia Patranjan (PaulaP)
Said Tabet (Said_Tabet)
Ugo Corda (Ugo_Corda)
Christian de Sainte Marie (csma)
Holger Lausen (holger)
Jos de Bruijn (JosDeBruijn)
Sandro Hawke (sandro)
Vassilis Tzouvaras (vassilis)
Paul Vincent (PaulVincent)
Guizhen Yang (Guizhen_Yang)
Axel Polleres (Axel_Polleres)
Chris Menzel (ChrisMenzel)
Mala Mehrotra (Mala)
Regrets (cf. http://www.w3.org/2005/rules/wg/wiki/2006-01-24_Meeting)
Dieter Fensel
Benjamin Grosof
Chair: Cristopher Welty
Scribe: Francois Bry
Agenda: http://www.w3.org/2005/rules/wg/wiki/2006-01-24_Meeting
http://lists.w3.org/Archives/Public/public-rif-wg/2006Jan/0085.html
IRC Log: http://www.w3.org/2006/01/24-rif-irc
RSS Agents Minutes: http://www.w3.org/2006/01/24-rif-minutes.html
Meeting Wiki Page: http://www.w3.org/2005/rules/wg/wiki/2006-01-24_Meeting
Summary of Topics
1. ADMIN (20 min)
2. Liaison (10 min)
3. Use Case & Requirements (20 min)
4. OWL & RDF Compatibility (15 min)
5. Classification (15 min)
6. AOB (5 min)
Summary of New Action Items
[NEW] ACTION: 1 to done.
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action01]
[NEW] ACTION: Allen to move the "importing rules to check data
compliance" use
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action03]
[NEW] ACTION: csma to ask ISO whether liaison is worthwhile for ISO IEC
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action02]
[NEW] ACTION: Donald will submit an email about the use case for
interchanging
rules specified in different metamodels
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action15]
[NEW] ACTION: JosB create a wiki page explaining the issue with bNode
semantics and summarize the possible solutions which have come up during
the
discussions on the mailing list
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action12]
same as
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action14]
[NEW] ACTION: JosDeBruijn create a wiki page explaining the issue with
bNode
semantics and summarize the possible solutions which have come up during
the
discussions on the mailing list
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action14]
[NEW] ACTION: JosDeRoo to update RIF wrt SPARQL rdfSemantics issue and its
pending resolution
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action13]
[NEW] ACTION: Said to start the discussion about a general use case for
rule
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action04]
Detailed Minutes
1. ADMIN
Next meeting: 31 Jan 2006
Minute last meeting extended with 2 modifications:
1. clarifies attendees list.
2. Persons to be added to the attendee list (of the telecon of 17 Jan)
should send an email to Christian de Sainte Marie.
No amendment to agenda
2. Liaison
Christian de Sainte Marie has clarified with Elisa Kendall whom she
meant I should contact (and that is Ed Barkmeyer). I will be able to
do the action up till next week re finding out whether a liaison with
that WG would be worthwhile (see recording action 2 as continued,
below).
liaison report from OMG SBVR - Allen Ginsberg:
1. SBVR current liaison work is with proposed "Interchange of
Human-oriented Business Rules" general use case. 2. Final Adopted
Specification of OMG "Semantics of Business Vocabulary and Busines
Rules" will be published in a few weeks.
liaison report from PRR (OMG) - Paul Vincent:
reported that PRR submitted a revised draft on Jan. 23. Somebody
(Sandro, probably) said that, thanks to an agreement with OMG, W3C
member had access to OMG WIP (and, thus, to that revised draft) even if
they were not OMG member. Somebody said, as a conclusion, that he would
send out something: could be Paul saying that he would circulate the URL
for the document or Sandro saying that he would circulate the info on
how W3C member can access OMG documents.
http://www.omg.org/cgi-bin/doc?bmi/06-01-07 for retrieving the latest
PRR submission.
No liaison reports from SPARQL (W3C) - Enrico Franconi,
XQuery, XPath (W3C) - Massimo Marchiori,
ODM (OMG) - Elisa Kendall
3. Use Case & Requirements
Chris Welty suggest extends one wek deadline for action 3.
Christian de SAinte Marie: goal was to make people look at use cases.
Said Tabet: I suggest deadline extension
Frank McCabe: Perhaps reviewing the use cases should be a standing
action on all
Paula-Lavinia Patranjan: +1 for deadline extension
Chrisitan de Sainte Marie suggests extend deadline and send email all
for informing of deadline extension.
Igor Mozetic: +1
Leora Morgenstern: +1 for deadline extension
Jeff Pan: +1
Francois Bry: +1
Giorgos Stamou: +1
JeremyCarroll: +1 to sandro
Sandro Hawke: what happens to feedback that's not in by the deadline?
Chris Welty suggests deadline extended by another week every one asked
to look at general use cases.
Jeremy Carroll: W3C process requires all comments received before the LC
deadline for comments to be addressed before CR.
Christian de Sainte Marie: feedback not in by this deadline wont go into
first working draft
Christian de Sainte Marie: feedback needed at latest next week if draft
expected for next f2f.
Jeremy Carroll: that seems to be the background to late comments can only
be ignored after that LC deadline
Chris Welty Decision: one more week to go over use cases and ensure we
cover all use cases we want to mention in the first working draft.
next ACTION: Ian to write scenario for Richt KR.
Allen Ginsberg: something has been done.
Chris Welty: if someone has finished an action, eg on the wiki, please
send message to chairs.
Chris Welty: new scenario by Christian for publication usecase.
Christian de Sainte Marie: there might be more than one use case here.
Is being further discussed on the mailing list.
Christian de Sainte Marie: As I look for the substance
("substantifique moelle" after Rabelais
http://www.kirjasto.sci.fi/rabela.htm)
of this use case, this reminds me of a discussion that maybe use case could
be looked at from other viewpoints, or dimensions. I'll care for starting a
discussion on that on the mailing list.
Chris Welty: "Interoperability between rule engines" Not yet done.
Leora Morgenstern: Decision Support detailed scenario action under progress.
Chris Welty: Allen and David about UC?
Chris Welty: More UC issues to discuss?
Allen Ginsberg or David Hirtle: It is time to look at requirements.
Allen Ginsberg or David Hirtle: New UC annonced.
Paula-Lavinia Patranjan: "my" action (detailed scenario for trust
establishment) not fully done. Input from Benjamin missing.
Said Tabet offers to anser in lieu of Benjamin.
JeremyCarroll: asks if human oriented rules are out of scope.
Gary Hallmark: +1 for out of scope
Donald Chapin: General Use Case on quality assurance.
Said Tabet: let us continue discussion by email. Applications needed for
standard being used.
Christian de Sainte Marie: looking at publication use case, target is
human and machine, RIF needed for being formal.
Christian de Sainte Marie: formal, ie unambigous and therefore machine
processable.
Christian de Sainte Marie: human oriented rules seem to me to mean "
informal" or less formal.
Donald Chapin: Use Case is not about human communication, and should not be.
Ian Horrocks: If people are using RIF that way, then it is fine. The same
rule can be interpretated differently.
Allen Ginsberg: depending obn culture or legal situation, etc.
Donald Chapin: The Human-oriented Rules Use case is not about human
processing
of rules, nor humans communicating with rule systems. It is about formal
exchanges between rule management systems that communicate with humans.
For
example, when organizations with different rule systems have to work
together,
their systems need to exchange business rules without losing semantics,
even
if some of those rules are implemented by humans.
Chris Welty: Interchange between rule systems with same or different
meta-models?
Chris Welty: Said's message starting this thread:
http://lists.w3.org/Archives/Public/public-rif-wg/2006Jan/0088.html
Sandro Hawke: Donald Chapin, maybe you can clarify what you just said for
the scribe, when you get the chance.
Chris Welty: We have to make sure that we can exchange rules that have a
formal semantics. Use Case must make clear that RIF only supports
applications.
Gary Hallmark: want to exchange rules that can be machine edited but not
machine executed?
Donald Chapin: Distinction needed: what is supported, what not.
Chris Welty: It should be a use case, not a requirement.
Chris Welty: Adding modality to rules for human communications.
Frank McCabe: deontic operators may well show up in compliance handling.
Donald Chapin: modal operators needed.
Christian de Sainte Marie: If there is such a requirement, it should be on
interchanging rules specific to communication between humans.
Donald Chapin: Requirement must not be on the RIF.
Allen Ginsberg: ???
Evan Wallace: I thought we were only talking about use cases at this point.
Requirements analysis comes in next stage.
Christian de Sainte Marie: use cases are about requirements, aren't they
Allen Ginsberg: Natural language might make sense for human communication.
Maybe it is only about human understanding.
Said Tabet: we are not adding more requirements.
Giorgos Stamou: maybe this use case strenghts some requirements
Gary Hallmark: maybe the use case is collaborative rule developement?
Christian de Sainte Marie: collaborative rule devt is covered by the
generalized use case nr 3.
Ed Barkmeyer: To me there are two issues from this use case:
(1) whether we need to support modals. There are several models for modals
that we could use if we need to.
(2) whether we need to support "non-machine-interpretable actions" in
rules.
Non-machine-interpretable conditions can look like External Functions.
Harold Boley: we should focus on UC before coming to requirements.
Human-orientedness good for advertising. We should work out this new
use case.
Jeremy Carroll: note that we may *reject* this use case during review
Jeremy: this use cazse might not go into working draft.
... there are dividing opinions on this use case.
Chris Welty: no decision yet. We should not reject Use Case before we see
it. We should understand what requirement this new use case would imply.
John Hall: We should discuss this use case /matter on email.
Chris Welty: agreed.
Christian de Sainte Marie: we do not want reject intersting use case.
Chris Welty: decsion: authors write down this use case so as we
understand it.
4. OWL & RDF Compatibility
Chris Welty: Email thread about semantics of RDF bnodes and relation to
SPARQL.
Christian de Sainte Marie: using SPARQL and in general query languages
might
be a good way of ensuring compatibility with RDF and other languages, at
least on the body/query side. I know people who supported it.
SPARQL and in general query languages on the
query/body side seem to be genrally accepted. What about Query on the
consequence side?
Sandro Hawke: I wonder if SPARQL doesn't give us a good Condition part of a
rule.
Christian de Sainte Marie: SPARQL query as body would be a good
compatibility
with RDF. Why not the same approach on the conclusion side? If we use query
languages as a mean to ensure compatibilty on the query side, why not also
on the conclusion side?
Jos De Bruijn: This would an rule extension of a query language. Like eg
Datalog.
Francois Bry: RIF should not be defined as extension of one or several
query
langhuages. This would be too complicated, take more time than
available, and
be rather risky (concerning eg semantics). Queries expressed in one or
several query languages could be allowed in rule bodies thru procedural
attachement. Queries do not belong in conclusion side, they are no data
def,
but queries.
Jos De Bruijn: +1 do not extend query language
Christian de Sainte Marie: +1 do not extend QL, my point was to start the
discussion on the conclusion side, knowing that queries can be used on the
condition side
Gary Hallmark: Oracle found that a subset of SQL is useful in ECA rules
Harold Boley: in rule bodies queries from different languages, including
SPARQL, can be called as external predicates; eg, SWRL calls OWL predicates
from RuleML rules.
Jos De Bruijn: SWRL allows queries in rules?
Harold Boley: SPARQL templates could be represented as (interpreted)
functions defined via rules having an equation in the head.
Edward Barkmeyer: my concern is that interpreting RDF like SPARQL
may not be compatible with the RIF semantics. So we can't just include
SPARQL syntax and retain its semantics.
Jos De Bruijn: SPARQL does not adhere to RDF semantics
Edward Barkmeyer: SPARQL inferences in rule heads may not be compatible
with
the RIF model.
Axel Polleres: hmmm, if we allow SPARQL (or any other query language) in
the
body and the query is recursively dependent on the rule consequent... we
are
gonna run into some issues
Edward Barkmeyer: merging RDF facts with rule engine facts will yield
interesting quertions on what the interpretation migft be.
Chris Welty: This is a general problem, not restricted to Query languages.
Jos De Bruijn: being member of SPARQL WG: SPARQL query is noly filter.
Therefore can be used in rule bodies. We do not see any issue at all if
looking at SPARQL as filter rule.
Chris Welty: How to handle, meaning of query language in consequents
unclear.
Chris Welty: Michael Kifer should now speak.
<Michael Kifer disconnected.>
Chris Welty: Further comments on that point?
Chris Welty: Bnode semantics?
Jos De Bruijn: we cannot resolve it yet
Chris Welty: Email exhange look like it has come to a resolution. Anyone to
write down on that?
HaroldBoley: Besides the antecedent and an ordinary consequent, there may
also be an equality consequent, where the equation right-hand side could be
the template of SPARQL.
Michael Kifer: We cannot make any clear decision wrt semantics. We can
write
down in rthe wik what the issues/problems are.
Jos De Brujin: Otherwise we would copmmit the RIF.
Jeremy Carroll: Can we have e-mail to notify us after wiki update
Michale will write in wiki issues related to bnode semantics.
Jos De Brujin will update wiki page on bnode semantics.
Jos De Brujin: Comment on SPARQL: seems to also disregard bnodes.
Chris Welty: Volunteer to write about bnode/SPARQL?
Jos De Roo: will report on what is going on in SPARQL concerning bnodes.
Someone: Enrico is also in SPARQL WG and could/should also
report on node issue.
Jos De Bruijn: ACTION: JosB create a wiki page explaining the issue with
bNode semantics and summarize the possible solutions which have come up
during the discussions on the mailing list
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action12]
5. Classification
Chris Welty:
http://www.w3.org/2005/rules/wg/wiki/Rulesystem_Arrangement_Framework
6. AOB
JosDeRoo: ACTION: JosDeRoo to update RIF wrt SPARQL rdfSemantics issue and
its pending resolution
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action13]
Sandro Hawke: ACTION: JosDeBruijn create a wiki page explaining the issue
with bNode semantics and summarize the possible solutions which have
come up
during the discussions on the mailing list
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action14]
Donald Chapin: action Christian was going to do about OWL/RDF compatibility
discussions. What to do with rules based on two meta-models?
Donald Chapin: If meta-models are different, you cannot exchange the rules.
We need use case stating what to do with different meta-models.
Said Tabet: +1
Edward Barkmeyer: +1
Donald Chapin: ACTION: Donald will submit an email about the use case for
interchanging rules specified in different metamodels
[recorded in http://www.w3.org/2006/01/24-rif-minutes.html#action15]
Received on Friday, 27 January 2006 12:08:44 UTC