W3C

- DRAFT -

RIF telecon 5 February 2008

05 Feb 2008

Agenda

See also: IRC log

Attendees

Present
csma, Sandro, ChrisW, hassan, Harold, AxelPolleres, StellaMitchell, IgorMozetic, +1.631.833.aaaa, MichaelKifer, Gary_Hallmark
Regrets
DaveReynolds, PaulaLaviniaPatranjan, Fran�oisBry, LeoraMorgenstern, PaulVincent
Chair
Christian de Sainte Marie
Scribe
Harold

Contents


 

 

<ChrisW> zakim harold is hassan

<sandro> oops

<ChrisW> Scribe: Harold

<AxelPolleres> josb, I would refer you to elisa for such conversations, alternatively! ;-)

<ChrisW> PROPOSED: accept minutes of telecon January 29

<ChrisW> RESOLVED: accept minutes of telecon January 29

<AxelPolleres> 408 done.

407 on Jos: Done

402 on Sandro: Done

Chris, can you enter your Continued?

373 on Sandro: TBD by Feb 8.

292 on Sandro: Waiting for other doc to settle done.

274 on Paula: sent regrets.

152 on Vincent: sent regrets.

Liason

<josb> I will dial in again

OWL:

Peter responded to Chris' response.

(Jan 29th)

F2F

<csma> http://www.w3.org/2002/09/wbs/38457/f2f9/

OWL

Agreed on a solution.

Have a draft.

Will send to WG if everyone agrees.

In two weeks.

(all of the above by Jos)

Chris: Weekend before f2f?

Jos: Yes, what we'll have by then.

Feb 17th.

<ChrisW> ACTION: debruij2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action01]

<trackbot-ng> Sorry, couldn't find user - debruij2

<ChrisW> ACTION: jdebrui2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action02]

<trackbot-ng> Sorry, couldn't find user - jdebrui2

<ChrisW> ACTION: jdebruij2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action03]

<trackbot-ng> Created ACTION-409 - Produce first draft of OWL proposal by Feb 17 [on Jos de Bruijn - due 2008-02-12].

F2F9

<csma> http://www.w3.org/2002/09/wbs/38457/f2f9/

Christian: Plse fill out that questionnaire asap.
... also if you cant attend.

<josb> all fine

F2F10

Christian: Should start thinking where it should be in May.

<ChrisW> f2f10 should be hosted by someone willing to throw a party

<sandro> heh

Christian: If you want to host it, put it into the wiki: When exactly, etc.

<josb> Axel, Igor?

Axel: Can check back, yes.

Christian: Preference would be in Europe, to keep balance, but not a necessity.

<AxelPolleres> I will check back with our admins, whether May is good/ok to host f2f.

<IgorMozetic> I can check for F2F10 as well

Michael: How to get from the airport to hotel.s

Christian: Depends on arrival time.
... Place d'Italie: Public transportation.
... Cab would be about 70 EUR.
... but around noon (any rush hour) would take longer.

Public transportation would take approx. 1 hour.

<ChrisW> ACTION: christian to post directions to hotels [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action04]

<trackbot-ng> Created ACTION-410 - Post directions to hotels [on Christian de Sainte Marie - due 2008-02-12].

New BLD and FLD

Christian: Changed link to new BLD.

MichaelK: Basically, mainly rewritten from almost scratch.

FLD: Extensibility framework (50-60%)
... NAF, etc. prepared.
... Syntax and Semantics.
... BLD then defined as dialect derived from this framework.
... Syntax (1st section) starts with parameters to define the dialect.

<Hassan> Why am I not listed (as Hassan)?

FLD: Rest, without ref to framework.
... Skipping 1st section, you dont need to read framework.

Semantics: how to derive BLD semantics from that of FLD.
... Save a few paragraphs.

<AdrianP> Hassan, I also have problems with zakim

XML still missing.

Chris: Schedule? Can we freeze? For f2f.

Michael: Yes, syntax.
... Semantics, by Fri.
... FLD finished, so derived Semantics of BLD.

Christian: Why not include negation into BLD?

Michael: Different kinds of negation (stable, well-founded, ...), so takes time.

Christian: Since BLD is not meant to be minimal, we could add one later.

Chris: That's the point where dialect splits happen, so not so easy.

Michael: Strong negation, perhaps, but still not good to hardcode into BLD.

<AxelPolleres> http://www.w3.org/2005/rules/wg/wiki/negation ... different forms of negation and why they don't fit together so well.

Chris: Let's get reviewers.

FLD now.

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

Michael: All in the OLD (Moin Moin) wiki!

Harold: Let's not change the tools while we are running :-)

<ChrisW> ACTION: cwelty to review FLD [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action05]

<trackbot-ng> Created ACTION-411 - Review FLD [on Christopher Welty - due 2008-02-12].

<AxelPolleres> by f2f only, I am afraid.

Jos: Will review BLD by the F2F but not before the next telecon.

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

Igor: Cannot promise reviewing new BLD by Tuesday, but try it by next week (Wed).

Christian: wiki-tr did not work for new BLD, so disconnected it.

Sandro: Format is wrong.

Christian: Because of HTML?

Sandro: Formatting issues could be debugged in a few minutes.

<ChrisW> ACTION: igor to review bld by F2F [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action06]

<trackbot-ng> Created ACTION-412 - Review bld by F2F [on Igor Mozetic - due 2008-02-12].

Christian: Frozen versions by Fri, Feb 15.

Lists

http://www.w3.org/2005/rules/wg/wiki/Core/List_Constructor

Harold: Removed Pair from syntax -- it's only a semantic means now.

Jos: No chance to read in detail, but looks fine.

Igor: Had little chance to read in detail, but looks fine.

<Hassan> I cannot be heard!

<Hassan> Who am I?

<sandro> zakm, unmute hassan

Harold: One issue is that Seq is polyadic (can be called with 0, 1, 2, or more arguments). For that we can instantiate FLD to BLD in a way that allows polyadic functions, namely Seq.

Hassan: Dont understand Ex. 3.
... Syntactic unific. seems to need pairs.

Harold: They are in the interpretation.

Jos: Propose to clarify by removing mentioning of unification.

Hassan: Agree.

Lists can be decomposed, e.g. in conditions using explicit syntactic equality (unification) as shown in Example 3.

-->

<Hassan> I will read the result

Lists can be decomposed, e.g. in conditions using explicit equality as shown in Example 3.

Sandro: wiki-tr now works properly on the new BLD.

Christian: Wonderful.

<csma> http://burns.w3.org/cgi-bin/wiki_tr?source=http%3A%2F%2Fwww.w3.org%2F2005%2Frules%2Fwg%2Fwiki%2FBLD&Go=Go

Michael/Sandro: some garbage left, but will go away.

Metadata

<josb> http://www.w3.org/2005/rules/wiki/Annotations

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

Jos: Also want to tackle Directives.
... So call the superclass 'Annotations'
... Metadata like Author, Date, ... can be ignored.
... Directives like DaveR's import ... cannot be ignored.
... For email exchange with DaveR: Probably want structure for Metadata.
... Inspired by Turtle syntax of RDF.
... Mapping from XML to RDF.
... Shows how RDF processors could use it.

Christian: In the RDF Compatibility doc.

Jos: Yes, later.

Michael: What is Metadata Statement? No example of this.

<josb> MetadataStatement ( a dc:title "bla" )

Jos: Would be Statement about arbitrary IRIs.
... A possible extension.

Christian: Also a bit skeptical about arbitrary IRIs.

Michael: Who must understand what Import means.

Chris: The dialect.

Christian: Difference between Metadata and Directives.

<Zakim> sandro, you wanted to ask why directives are grouped with metadata

Sandro: Dont understand why directives are not fully part of the language?

Jos: You dont need to change the language when you change the directives.

Michael: Like include statement in C (part of preprocessor)

Sandro: So we would need two processors.
... all directives are processed before the rules are processed.

Chris: Only by analogy.

Sandro: The model-theoretic semantics is not the differentiating criterion: Also part of PRD is not specified model-theoretically.
... Gray line between the language itself and its directives.
... Illegal directives should be caught by schema checker.

Christian: Could be included as a piece of syntax rather than as a directive.
... We could say we have no directives, so things get easier.

<sandro> Sandro: I am comfortable with the idea that directives are style-decision -- you never need to use them, but maybe they will be the better style for some kinds of things.

Jos: Then it would be harder to make extensions.

<sandro> Sandro: I don't think extending the syntax is hard.

Chris/Christian: Agree with Sandro.

Christian: Why changed presentation syntax?

Jos: If we want to specify anything, it must go there.

Christian: Maybe not call it 'Annotation' to avoid confusion with XML Schema Annotations.

Jos: More worried with OWL Annotations.

Sandro: In XML Schema all Annotations are optional.
... so confusion possible here.

Jos: OK when better name can be found, but first decide about 'Directives'

Christian: Strawpoll.

<ChrisW> Favor Two kinds?

<ChrisW> +1

<Hassan> 0

<AxelPolleres> 0

<josb> +1

<IgorMozetic> +1 metadata+directives

+1

<MichaelKifer> +0

<sandro> -0.5

<GaryHallmark> why not just define the specific directives you want. the only examples I've heard are import/include

Christian: Yes, Gary, it may be in the syntax of the language itself.

<ChrisW> Favor one kind of metadata

<GaryHallmark> +1

<sandro> +1

<AxelPolleres> 0

<Hassan> 0

<IgorMozetic> -0

<josb> 0

0

<ChrisW> 0

<MichaelKifer> -0

<Hassan> I need to be convinced of the need

Sandro: Want to do it on a case by case basis.
... Import should be part of the language.

Chris: What diff does it make?

Sandro: Avoid unnecessary complication.
... Schema validation can help.

<GaryHallmark> are there any directives other than import?

Harold: Schematron can always be used for finer-grained validation.

<josb> Suggested directives for BLD: rif:imports, rif:requiresDataSet, rif:dataModel

<sandro> Sandro: I'm not opposed to directives in principle, it just seems unnecessary for the examples I'm seeing here.

<GaryHallmark> jos, what do the last 2 do?

Christian: Next step would be to list all RIF directives of interest.

Sandro: Having directives does change model theory.

<Hassan> why not? imports are theories...

Harold: Model theory kicks in only after import.

<josb> ref to dataset and data model is data model of the rules

<josb> see http://www.w3.org/2005/rules/wg/wiki/Arch/Data_Sets

Christian/Jos: If we had modules, things would be different.

Jos: Then would not use directives to refer to modules, but use syntax in the language.

Chris: and semantics.

Christian: Was referring to datasets, not modules.

Chris: Wrt model theory, you treat them (datasets?) as if they were part of the symbol space.

Christian: If you want to cross-ref between datasets, how do you do that?
... If this dosent change the model theory, you cannot see it as preprocessing.

Chris: Lost me here.

Christian: Will make up an example.

<ChrisW> +1

<IgorMozetic> +1

<csma> +1

<josb> +1

<sandro> -1 don't know enough yet

<Hassan> 0

0

<AxelPolleres> +1

<ChrisW> previous poll about prefering Jos' metadata proposal

<GaryHallmark> -0.5

<Hassan> +1 on adjourning

<GaryHallmark> for interop, it would be better to import a rif document containing ground facts than to use rif:requiresDataSet

<IgorMozetic> +1, bye

<ChrisW> ACTION: Christian to write up an example of what he means regarding the metadata proposal [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action07]

<trackbot-ng> Created ACTION-413 - Write up an example of what he means regarding the metadata proposal [on Christian de Sainte Marie - due 2008-02-12].

<AxelPolleres> actions to igor and myself to check back f2f10 options...

<AxelPolleres> bye

Summary of Action Items

[NEW] ACTION: christian to post directions to hotels [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action04]
[NEW] ACTION: Christian to write up an example of what he means regarding the metadata proposal [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action07]
[NEW] ACTION: cwelty to review FLD [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action05]
[NEW] ACTION: debruij2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action01]
[NEW] ACTION: igor to review bld by F2F [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action06]
[NEW] ACTION: jdebrui2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action02]
[NEW] ACTION: jdebruij2 to produce first draft of OWL proposal by Feb 17 [recorded in http://www.w3.org/2008/02/05-rif-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/02/05 17:31:26 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/10/17/
Succeeded: s/allows on polyadic function, namele Seq./allows polyadic functions, namely Seq./
Found Scribe: Harold
Inferring ScribeNick: Harold
Default Present: csma, Sandro, ChrisW, hassan, Harold, AxelPolleres, StellaMitchell, IgorMozetic, +1.631.833.aaaa, MichaelKifer, Gary_Hallmark
Present: csma Sandro ChrisW hassan Harold AxelPolleres StellaMitchell IgorMozetic +1.631.833.aaaa MichaelKifer Gary_Hallmark

WARNING: Replacing previous Regrets list. (Old list: DaveReynolds, PaulaLaviniaPatranjan, Fran�oisBry, Leora_Morgenstern, PaulVincent)
Use 'Regrets+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Regrets+ DaveReynolds, PaulaLaviniaPatranjan, Fran�oisBry, LeoraMorgenstern, PaulVincent

Regrets: DaveReynolds PaulaLaviniaPatranjan Fran�oisBry LeoraMorgenstern PaulVincent
Agenda: http://lists.w3.org/Archives/Public/public-rif-wg/2008Feb/0010.html
Got date from IRC log name: 05 Feb 2008
Guessing minutes URL: http://www.w3.org/2008/02/05-rif-minutes.html
People with action items: christian cwelty debruij2 igor jdebrui2 jdebruij2

[End of scribe.perl diagnostic output]