RDB2RDF Working Group Teleconference -- 10 Jan 2012

<trackbot> Date: 10 January 2012

<mhausenblas> trackbot, start telecon

<trackbot> Meeting: RDB2RDF Working Group Teleconference

<trackbot> Date: 10 January 2012

<juansequeda> zakime, aabb is me

<mhausenblas> scribenick: Ashok

Admin

<mhausenblas> PROPOSAL: Accept the minutes of last meeting http://www.w3.org/2011/12/20-RDB2RDF-minutes.html

RESOLUTION: Minutes accepted as posted ... no objections

ISSUE-68: Multiple PredicateMaps in a PredicateObjectMap

<mhausenblas> ISSUE-68?

<trackbot> ISSUE-68 -- Multiple PredicateMaps in a PredicateObjectMap -- open

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/issues/68

<MacTed> might make sense to start http://www.w3.org/2001/sw/rdb2rdf/track/actions/open

Richard summarizes the current situation

<Souri> PO+ is more common, then why not PO+ as well? Why just P+O?

<dmcneil> +q

David: Let us focus on getting the core spec out ... shortcuts can come later

<Souri> There was NO resolution on it!!

<Souri> I was NOT happy to see it in the pre-LC draft w/o any resolution!

Richard: Many D2R users use this feature

Souri: PO+ is more common, then why not PO+ as well? Why just P+O?

<dmcneil> richard: out of curiousity... why are you opposed to supporting multiple objects on a PO map? Is it simply because you don't see it as a very common need?

Richard: Very useful to create multiple propertes from one column ... other wise need to copy/paste

<Souri> Leaving out the shortcut => x:omap rr:objectMap [ .... ] ; [] rr:predicateObjectMap [ rr:predicate x:p1 ; rr:objectMap x:omap ], [ rr:predicate x:p2 ; rr:objectMap x:omap ] .

Ted: Has there been a user need for multiple object maps. Or is it just symmetry

Souri: This feature was suggested after Last Call
... it was added to document without a WG resolution
... PO* is more common ... not just a matter of symmetry

<mhausenblas> Michael: I'd very much appreciate it if people would stop arguing about process issues. Things that happened, happened. We have overcome more difficult issues already.

<mhausenblas> Michael: Otherwise, my proposal would be to go into a 2nd LC

Souri: User do not know that PO+ is even possible
... Language has been changed since D2R

<juansequeda> ... haven't made my mind yet. I agree with both parties

<cygri> PROPOSAL: allow multiple predicate maps on a predicate-object map

<Souri> Three proposals: none, P+O only, P+O and PO+ both

<dmcneil> richard: out of curiousity... why are you opposed to supporting multiple objects on a PO map? Is it simply because you don't see it as a very common need?

<Souri> -1

<dmcneil> +q

<Zakim> cygri, you wanted to say it was *not* suggested after LC

<Souri> I always stick to the fact

Ashok: Let us ask who can live with the proposal

<mhausenblas> Michael: If we ask only who can live with the proposal, would that mean that people who don't say anything, agree with it?

Richard: We could add a number of shortcuts but we should add only those where there is a demand

<dmcneil> so in D2R it is easy to add multiple objects?

<Souri> Example of PO+: Three cols combined into one predicate: cellphone, workphone, homephone

David: It is easy for people to add an object ... or a predicate?

Richard: Typically D2R users do a default generation and then specialize it ... can add additional predicates

<Souri> PROPOSAL: allow no shortcuts regarding PredicateMaps and ObjectMaps, only P+O shortcut, or both P+O and PO+ shortcuts

<MacTed> straw poll: would you object to spec including (1) neither, (2) P+O only, (3) P+O and PO+ both

<juansequeda> (3) +1

<Souri> let us make the proposal +ve

<MacTed> straw poll: would you prefer spec include (1) neither, (2) P+O only, (3) P+O and PO+ both

<Souri> (1) +1, (2) -1, (3) 0.5

<juansequeda> (3) +1

<joerg> (3) +1

<nunolopes> (1) +1, (2) -1, (3) +1

<juansequeda> (1) -1

<juansequeda> (2) -1

<dmcneil> (1) +1

<cygri> (1) -1, (2) +1, (3) -0.sometihng

<seema> (1) +1, (2) -1, (3) +1

<ivan> (1) -1 (2) 1 (3) 1

<Souri> (3) is the winner

<Souri> it is not an argument for symmetry!

<Souri> that impl difficulty note has to apply to both P+O and PO+

<mhausenblas> Michael: So, can we have a 'final' proposal now?

<dmcneil> +q

<MacTed> working....

<cygri> dmcneil, true. i don't believe that anyone will write R2RML mappings from scratch.

<MacTed> PROPOSAL: allow P+O and PO+, with language stating that either may be at risk if implementation is proven excessively difficult

<mhausenblas> PROPOSAL: Close ISSUE-68 with - allow P+O and PO+, with language stating that either may be at risk if implementation is proven excessively difficult

<mhausenblas> Michael: Anyone objects (you can also abstain)?

<Souri> "should"

<juansequeda> +1 to the Proposal

<Souri> -1 considering how would we assess the relative difficulties of the two shortcuts -- if P+O gets in PO+ must be in too OR both must be out

<cygri> PROPOSAL: Close ISSUE-68 with - allow P+O and PO+

<juansequeda> +1

<ivan> +1

<joerg> +1

<nunolopes> +1

<Souri> +1

<MacTed> +1

<seema> +1 to Richard's proposal

<cygri> +1

RESOLUTION: Issue 68 closed with Richard's proposal diuscussion - allow P+O and PO+

<mhausenblas> ACTION: Richard to implement the resolution to ISSUE-68 [recorded in http://www.w3.org/2012/01/10-RDB2RDF-minutes.html#action01]

<trackbot> Created ACTION-183 - Implement the resolution to ISSUE-68 [on Richard Cyganiak - due 2012-01-17].

<cygri> ACTION-182?

<trackbot> ACTION-182 -- Richard Cyganiak to email the group summarizing ISSUE-68 position -- due 2011-12-27 -- OPEN

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/182

<cygri> trackbot, close ACTION-182

<trackbot> ACTION-182 Email the group summarizing ISSUE-68 position closed

ISSUE-73: Section 11 cleanup

<mhausenblas> ISSUE-73?

<trackbot> ISSUE-73 -- Section 11 cleanup -- open

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/issues/73

<mhausenblas> PROPOSAL: Close ISSUE 73 as it was blocked on 68 and 72

RESOLUTION: ISSUE-73 closed as issues 68 and 72 resolved

<cygri> +1

<MacTed> need to convert ISSUE 73 to ACTION...

Towards CR

<mhausenblas> Michael: Please http://www.w3.org/2005/10/Process-20051014/tr.html#cfi

<mhausenblas> Michael: http://www.w3.org/2001/sw/rdb2rdf/wiki/Implementations

<mhausenblas> Michael: http://www.w3.org/2001/sw/rdb2rdf/wiki/Last_Call

Ivan: To go to CR we need a call ... for that we need a report on Last Call issue resolution

<ericP> http://www.w3.org/2001/sw/rdb2rdf/directMapping/LC/

Ashok: Is the DM document ready?

<mhausenblas> Michael: It seems to me that it may make more sense to wait till all R2RML changes have been implemented and do the CR resolution next week (?)

Eric: Talks about translating SQL strings ... will write some mail

Michael: So, the DM document is not ready

<cygri> see http://www.w3.org/2001/sw/rdb2rdf/r2rml/#sotd-xsd-dep

Ivan: Reminds the WG that XSD 1.1 will go to PR soon. So our issue on XSD 1.0 vs. XSD 1.1 becomes moot

<cygri> ACTION: cygri to remove XML Schema Datatypes Dependency paragraph from R2RML ED [recorded in http://www.w3.org/2012/01/10-RDB2RDF-minutes.html#action02]

<trackbot> Created ACTION-184 - Remove XML Schema Datatypes Dependency paragraph from R2RML ED [on Richard Cyganiak - due 2012-01-17].

Ashok: The issue was about the canonical representation

Souri: Examples with case sensitive table names would be good. Please check is current examples are correct wrt case sensitivity.

<mhausenblas> trackbot, end telecon

<MacTed> s/Examples with case sensitive table names would be good. Please check is current examples are correct wrt case sensitivity/Examples with table names with special characters and/or case sensitivity would be good. Please check that current case sensitive examples are correct/

- DRAFT -

RDB2RDF Working Group Teleconference

10 Jan 2012

Attendees

Contents

Admin

ISSUE-68: Multiple PredicateMaps in a PredicateObjectMap

ISSUE-73: Section 11 cleanup

Towards CR

Summary of Action Items

Scribe.perl diagnostic output