W3C home > Mailing lists > Public > public-rif-wg@w3.org > August 2006

[RIF] Draft minutes for August 29 2006 meeting

From: Alex Kozlenkov <alex.kozlenkov@betfair.com>
Date: Wed, 30 Aug 2006 22:46:49 +0100
Message-ID: <E39E50172D7A6546BCE3E981C985C9870BEDB738@UKMAIL.sportex.com>
To: <public-rif-wg@w3.org>
Guys,

Please find first draft of August 29th RIF meeting minutes. I am off to
holidays until September 18th so it may be needed for someone to clean
this up further before it is approved.

Thanks,

Alex Kozlenkov

Betfair Ltd.

  <http://www.w3.org/> 


- DRAFT -


SV_MEETING_TITLE


29 Aug 2006


See also: IRC log <http://www.w3.org/2006/08/29-rif-irc> 


Attendees


Present 

Regrets 

Chair 

SV_MEETING_CHAIR 

Scribe 

AlexKozlenkov 


Contents


*	Topics
<http://www.w3.org/2006/08/29-rif-minutes.html#agenda#agenda>  
*	Summary of Action Items
<http://www.w3.org/2006/08/29-rif-minutes.html#ActionSummary#ActionSumma
ry>  

________________________________

 

 

<csma> hello

<ChrisW> <ChrisW> tells alex how to scribe

<csma> scribenick: AlexKozlenkov

<ChrisW> next meeting next Tuesday

<ChrisW> 
http://lists.w3.org/Archives/Public/public-rif-wg/2006Aug/0021.html

<ChrisW> 
http://lists.w3.org/Archives/Public/public-rif-wg/2006Aug/0021.html

<ChrisW> the 8/8 meeting posted after SAid's amendments

<ChrisW> 
http://lists.w3.org/Archives/Public/public-rif-wg/2006Aug/0025.html

<ChrisW> no objections to accept 8/8 minutes

<AxelPolleres>  Wiki page for clarifying negation

<csma> ok, I do it

<ChrisW> ACTION: axel to make a wiki page for negation [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action01]

<AxelPolleres> it was entered without colon, thus not recognized.

<ChrisW> last week serious IRC problems
... accepted August 22 minutes
... liason activities--no outstanding actions
... PRR to be discussed this week?

<csma>  Fair Isaac have nominated somebody to represent here
... should be registered but I cannot see him

<ChrisW> still need a PRR liason

<AlexKozlenkov>  JBoss guys to join RIF, possible liason for PRR

Donald: successful Beijing meeting on SBVR
... ISO SBVR process successfully continued

<ChrisW> to discuss use cases now

allen: improve consistency
... issues 9 and 10 addressed
... could be changed back easily
... the look of the boxes not really a problem, not able to duplicate
issue 11
... could not see display problems
... working draft is OK in any browsers

sandro: the official one is fixed

allen: item 7 a more substantial one
... do this as a group, a separate teleconference perhaps

<ChrisW> we need to discuss issue 7
... issue 9 and 10 are done
... issue 11, we must keep track

sandro: if nobody objects, the issue will be closed

<ChrisW> editing and formatting 9 and 10, next week will be closed after
people have a chance to comment
... issue 11
... sandro, leave as an open issue

sandro: sure

<ChrisW> move it to a persistent action?

sandro: not sure

deborah: agree with that

<ChrisW> editors are working on the wikipage and the process is painful
for generating a working draft
... we should remeber as a group that a new draft is generated

sandro: everybody looks at the draft before it is published
... small changes can be made directly

<csma> Sandro has already an action on that (action ID 58)

<ChrisW> close issue 11

<scribe> ACTION: deborah close issue 11 [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action02]

<csma> ACTION: Deborah to close action item 11 (to be moved to action
58) [recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action03]

can we remove that?

<ChrisW> issue 7 is a review by Sven Groppe

<ChrisW> issue 7

<ChrisW> http://www.w3.org/2005/rules/wg/track/issues/7

<ChrisW> summarizing his views
... stylistic changes that I do not agree with

suggestion 1: start with the discussion of RIF and enumerate requiremens
rather than UC

<ChrisW> anyone agrees with this?
... I'd rather keep the document as is

sandro: let him have a look at the last draft

<csma>  he prefers to have UC linked to Requirements

<Hassan> it would not hurt to have a summary before each UC
... requirements are now present
... if we add links to requirements from UC, this would answer Sven's
concerns

<ChrisW> eventually, we will have these links

<Hassan> +1

<csma>  do not need to have examples of translation
... UC should make it clear that a translation is needed

<ChrisW> current UC's do that?

<csma>  not all
... use cases should not be about only about the use of rules but more
focused on interchange

<AxelPolleres>  we could go back to UC's and add concrete languages?
... it could be solved in this way

<ChrisW> 1st f2f decided that specific languages will not be mentioned

hassan: debated before, agree that the languages not required

<csma> +1 to Hassan not to have concrete language

<AxelPolleres>  mention an example?
... if we do not want this, so be it
... finally, it would be nice to have UC and requirements to be
connected

<ChrisW> this is the work to be done

<AxelPolleres>  still give example languages where this use case can
apply concretly (without referring to the concrete syntax) of languages

<csma> Is that clear that RIF is itself a rule language?

<Hassan> Good question, Chris!

allen: on interchange, interchange between RIF and language is also an
interchange
... RIF<->language is also an interchange

<ChrisW> the interchange should be specific, the wholistic view is not
enough

<AxelPolleres> actually, the original use caes were a lot more specific,
we abstracted them on purpose.

<csma>  to be more specific, I tried to extract the processing from UC
and failed
... that is a problem with UC non-specificity

<AxelPolleres> so, I suggest let's stick with it. However, it is true
that we need more concrete examples later on (when it comes to
implementations, etc.)

<Harold> Allen, you perhaps meant the term 'interchange' might already
be used for one-way translation from some existing language (eg. Prolog)
into the RIF (without translation out, maybe running it in RIF instead)?

<ChrisW> can christian make a more concrete statement about particular
UC's

<csma>  it looks that the processing model is missing

<Allen> Harold, I meant that the term could be used to cover that case
too

<ChrisW> ACTION: csma to post an example of processing model in use
cases [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action04]

<csma> ACTION: csma to post an example of an UC not being clear enough
about the processing model [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action05]

<ChrisW> another item from Sven's review: annotation of UC's to phase 1
or 2
... any comments on that?
... point 6 of review is now clear
... point 7: only exchange of the facts or rules for UC 1.1
... negotiating cross-rule contracts across platforms
... is it clear in the UC whether the rules or facts are interchanged
... seems reasonable
... it needs to be answered
... allen will have a look

<csma> about point 6, which way it has been clarified?

<ChrisW> RIF is about interchange
... the UCR document is now better, so it has been clarified in the
abstract of the 2nd WD

<Zakim> csma, you wanted to ask which way point 6 was clarified

<Harold> Do we regard facts as special rules (having 'empty-conjunction'
= 'true' bodies)? Or, do we treat (ground) facts specially (eg. to
access databases)?

<ChrisW> point 8: reorganize the document for separating UC and
requirements, it will become clearer

deborah: add that to the issue description

<ChrisW> there will overlap between UC and requirements but the
separation will be more clear

<csma> ACTION: Deborah to summarise the discussion to issue 7
description [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action06]

<ChrisW> rule extensions to OWL, will be done when requirements are
linked
... add conclusions
... references to resources for real-worlkd examples

<ChrisW> concrete languages or test cases
... test cases, we will have that, rule systems, no

<AxelPolleres> +1 to express as response the intention to add concrete
testcases to each use case later on.

<Hassan> eventually, we are converging to the XML dialect that will be
an intersection of a few languages

<ChrisW> it is needed but not in the UC
... the test cases is a separate document
... we start off with UC and requirements and that is how RIF is
progressing

<AxelPolleres> well, we should of course back-check whether the
testcases cover the use cases and vice-versa, right?

<csma> should we leave one week for the group to consider an issue?

<ChrisW> some issues could be shorter or longer

<csma> OK

<ChrisW> that finishes Allen's issues
... heartbeat requirements fot the next WD
... October 10, the next WD

<Hassan> +1 with csma

<ChrisW> a new draft of UCR not a problem?

allen: does not seem as big job

<ChrisW> the major part is associating UC and requirements

allen: is the link supposed to be complete?

<ChrisW> every UC should motivate a requirement
... is the opposite true?

<csma> For the record, what I said that Hassan supported is that we
should have the 1st WD of the tech spec as our objective for the next
heartbeat

sandro: it would be good

<ChrisW?> you wanted to muse about adding results of RIFRAF survey

sandro: use of XML, one example
... add results of RIFRAF questionnaire to the next UCR document
... just a suggestion

csma: the feedback will be useful but in UCR?

hassan: UCR is separate
... the classification work and technical design will use these results

<Harold> Sandro, I was disconnected from my Skype access to the W3C
bridge, and the conference code seems no longer to be valid.

<csma> answers should not be in the UCR document

<sandro> That's odd, Harold. Can you do the Admin Assistance code?

<Harold> Also *0 did not get me someone.

hassan: agree

<sandro> Ah.

csma: the feedback will be useful, however, so it will be useful to
publish that separately
... where?

<Harold> I used this.

<ChrisW> RIFRAF will be separate

<sandro> Very odd, Harold -- it just worked for me on a second line.

<Harold> OK, I'll try again.

hassan: produce an ontology of the RIFRAF

sandro: a non-normative result, not part of the standard

<Harold> Sandro, it works again, thanks!

csma: the answers should not be normative, not even a result
... requirements is the main thing

<ChrisW> the extensibility mechanism will be linked to that
classification

csma: got your point

sandro: the results of the classification may be part of the rectrack,
it is easier to remove than add

<ChrisW> certainly if we are publishing, we need editors for RIFRAF

<ChrisW> ACTION: on chairs to think aboutr RIFRAF editor [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action07]

<AxelPolleres> Depending that you concretize the role of the RIFRAF
document more... I could be a volunteer.

<ChrisW> sandro agree that a separate RIFRAF doc is required

csma: linking UC with requirements should be an issue

<ChrisW> what happend
... what happend

allen: no consistency across UC's

<csma> ACTION: allen to post the previous work on links between reqs and
UC [recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action08]

<ChrisW> allen any other issues in UCR?

<allen> no

<Zakim> csma, you wanted to propose a path to UCR WD3

csma: we should first agree on all short term issues
... for draft 3, we need to assign priorities

<ChrisW> for example, one week to submit the issues in the current WD

<csma> ACTION: chrsi to send email to inform everybody of process toward
UCR WD3 [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action09]

<ChrisW> within two weeks, the group agrees which issues will be
addressed with the WD3
... outstanding items
... for RIFRAF, outstanding actions

frank: simplify the questionnaire

hassan: the idea is to use a tree to open a branch
... in this case, one does not answer the lower level questions

frank: does the W3C questionnaire technology allow this

<AxelPolleres>  no it is not possible

<Hassan> The logical form could be followed so that semantic distance
could be computed

<Harold> Frank and Hassan, could we have cross-references between RDF(S)
and OWL compatibility on one hand and certain kinds of (order-sorted)
type systems on the other hand?

hassan: coould post the way it could be organized

<csma> ACTION: work with Frank to augment type discriminators proposal
(ID 88) [DONE] [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action10]

hassan: we will talk about it next week

frank: the questionnaire needs more structure

<ChrisW> Frank and Hassan could work on types, to make it more
hierarchical

<csma> ACTION: franck and hassan to work on a hierarchy of type-related
discriminators [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action11]

hassan: I started with types but it should be possible for all
discriminators

<AxelPolleres> we have to formalize which answers exclude each other and
then find an oprimally ordered BDD :-) then we have the ontology! :-)))

<sandro> sandro: does it make sense to do this in OWL now, or soon? is
it obvious to anyone how to do that?

<FrankMcCabe> That would be an interesting test case for OWL

<csma> ACTION: record questionnaire answers for JBoss and XUL in an
email (action 89) [CONTINUED] [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action12]

<csma> ACTION: sync questionaire questions back to RAF wiki page (action
90) [DONE] [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action13]

<csma> ACTION: include Paula's questions to questionnaire (action 91)
[CONTINUED] [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action14]

<ChrisW> action review is completed
... next week start talking about technical design

<sandro> +1 adjourn

<csma> +1

<PaulaP> bye

adjourn now

<JeffP> bye


Summary of Action Items


[NEW] ACTION: allen to post the previous work on links between reqs and
UC [recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action08]
[NEW] ACTION: axel to make a wiki page for negation [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action01]
[NEW] ACTION: chrsi to send email to inform everybody of process toward
UCR WD3 [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action09]
[NEW] ACTION: csma to post an example of an UC not being clear enough
about the processing model [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action05]
[NEW] ACTION: csma to post an example of processing model in use cases
[recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action04]
[NEW] ACTION: deboragh close issue 11 [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action02]
[NEW] ACTION: Deborah to close action item 11 (to be moved to action 58)
[recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action03]
[NEW] ACTION: Deborah to summarise the discussion to issue 7 description
[recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action06]
[NEW] ACTION: franck and hassan to work on a hierarchy of type-related
discriminators [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action11]
[NEW] ACTION: on chairs to think aboutr RIFRAF editor [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action07]
 
[PENDING] ACTION: include Paula's questions to questionnaire (action 91)
[recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action14]
[PENDING] ACTION: record questionnaire answers for JBoss and XUL in an
email (action 89) [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action12]
 
[DONE] ACTION: sync questionaire questions back to RAF wiki page (action
90) [recorded in http://www.w3.org/2006/08/29-rif-minutes.html#action13]
[DONE] ACTION: work with Frank to augment type discriminators proposal
(ID 88) [recorded in 
http://www.w3.org/2006/08/29-rif-minutes.html#action10]
 
[End of minutes]

________________________________

 




image001.gif
(image/gif attachment: image001.gif)

Received on Wednesday, 30 August 2006 21:54:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:33 GMT