<Ashok> scribenick: dmcneil

ashok: accept minutes of last meeting?

RESOLUTION: Accept minutes from June 21 meeting

ashok: next scheduled call is July 5, which might be difficult because of holiday

<Souri> +1 to cancel

ashok: should we cancel next week's telecon?

<juansequeda> +1 to cancel

<Seema> +1 to cancel

RESOLUTION: cancel July 5 telecon

<juansequeda> +1 from marcelo

ashok: we have two direct mapping implementations, will there be others from a commercial company as a product?
... do we have concensus that we do not need any more R2RML syntaxes? i.e. Turtle is enough

from the charter: "The mapping language SHOULD have a human-readable syntax as well as XML and RDF representations of the syntax for purposes of discovery and machine generation. "

PROPOSAL: R2RML will only define the Turtle syntax

<Ashok> ... and it meets the human-readable requirements

+q

souri: R2RML is defining an RDF model, as long as a mapping author writes an RDF document that corresponds to the RDF model, that is fine

<ericP> PROPOSAL: R2RML is defined in terms of an input RDF graph. For interoperability simplicity, implementations are SHOULD accept at least Turtle input.

<ericP> PROPOSAL: R2RML is defined in terms of an input RDF graph. For interoperability simplicity, implementations SHOULD accept at least Turtle input.

+1

<ericP> +1

<ericP> RESOLUTION: R2RML is defined in terms of an input RDF graph. For interoperability simplicity, implementations SHOULD accept at least Turtle input.

ashok: next issue is property names

<Souri> rr:subjectMap property rr:Subjectmap

<ericP> that's pretty common in RDF

i am having a hard time hearing

<Souri> rr:subjectMap property rr:SubjectMap

<ericP> http://www.w3.org/mid/20110623160019.GB30806@w3.org

<ericP> PROPOSED: R2RML property and classes are named per http://www.w3.org/mid/20110623160019.GB30806@w3.org 's "drop use" option

<ericP> +1

<boris> +1

<ericP> RESOLUTION: R2RML property and classes are named per http://www.w3.org/mid/20110623160019.GB30806@w3.org 's "drop use" option

<Ashok> ACTION Souri to implement names as resolved above

<trackbot> Created ACTION-136 - Implement names as resolved above [on Souripriya Das - due 2011-07-05].

<ericP> +1 to IRIs for term types

<MacTed> action 136 is going to need clarification... "resolved above" won't make sense when we look at it later.

<trackbot> Sorry, couldn't find user - 136

<boris> rr:termtype rdfs:range {"IRI", "BlankNode", "Literal"} .

<Souri> rr:termtype rdfs:range {"IRI", "BlankNode", "Literal"} .

discussion of ISSUE-45

souri can write up a proposal of IRIs to use

<Ashok> ACTION: Souri to create proposal for IRIs to resolve ISSUE-45 [recorded in http://www.w3.org/2011/06/28-rdb2rdf-minutes.html#action01]

<trackbot> Created ACTION-137 - Create proposal for IRIs to resolve ISSUE-45 [on Souripriya Das - due 2011-07-05].

<Souri> thinking of something simple like: rr:IRI, rr:BlankNode, rr:Literal

next is ISSUE-47

+q

<MacTed> issue 47?

<MacTed> issue-47?

<trackbot> Sorry... I don't know anything about this channel

<MacTed> trackbot, start meeting

<trackbot> Sorry... I don't know anything about this channel

<trackbot> If you want to associate this channel with an existing Tracker, please say 'trackbot, associate this channel with #channel' (where #channel is the name of default channel for the group)

<MacTed> trackbot, associate this channel with #RDB2RDF

<trackbot> Associating this channel with #RDB2RDF...

<trackbot> Sorry... I don't know anything about this channel

<trackbot> If you want to associate this channel with an existing Tracker, please say 'trackbot, associate this channel with #channel' (where #channel is the name of default channel for the group)

<MacTed> unhelpful, bot...

souri: the question is whether we automatically apply percent encoding to the data values used as IRIs
... would prefer to do this automatically because it simplifies it for the user
... we could refer to the standard for this

<ericP> i propose all characters outside of ([^<>"{}|^`\]-[#x00-#x20])* per http://www.w3.org/TR/rdf-sparql-query/#rIRI_REF

<Ashok> ... Percent encode all characters outside of ([^<>"{}|^`\]-[#x00-#x20])* per http://www.w3.org/TR/rdf-sparql-query/#rIRI_REF

ericP: still a question of whether to convert spaces to "+" instead of "%20"

<Souri> In Oracle utl_url.escape('http://www.acme.com/a url with space.html') Returns: http://foo.com/a%20url%20with%20space.html

<Souri> space => %20

<boris> most URL encoders convert space to %20

ashok: it is hard to make the argument that we should follow the standard approach with one exception

<MacTed> + is LONG deprecated as space substitute

<ericP> d2r has some functions near this

MacTed: using + for space is a long deprecated concept

we should not do anything to produce a +

ericP: where is it deprecated?

<ericP> weak!

<ericP> +1 to functions

dmcneil: how about as we discussed on the mailing list, to have the mapping explicitly invoke conversion functions

this removes the automatic behavior

but we have to specify the functions

<Souri> rr:template "xyz/{rr:foo(empno)}"

<MacTed> ericP - http://en.wikipedia.org/wiki/Percent-encoding#The_application.2Fx-www-form-urlencoded_type

<Souri> or more like: rr:template "xyz/{rr:foo({empno})}"

ashok: let's wait and discuss further in two weeks

next to discuss is ISSUE-22

next, ISSUE-36

ashok: Richard proposed we postpone this

+1

<Souri> +1

<boris> +1

<ericP> +1

ashok: i.e. the proposal is that we not define the schema in the mapping output

RESOLUTION: Resolve ISSUE-36 as POSTPONED. The initial version of R2RML will not provide special features for creating schema triples.

ISSUE-37

<boris> ?

ashok: why would we even consider doing this?

<juansequeda> what is issue-37?

<boris> ISSUE-37?

<trackbot> ISSUE-37 -- Allow specifying DB connection details in the R2RML mapping -- open

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/issues/37

MacTed: the reason is not to specify the connection so much as clarifying the data source

part of the map to the data includes where it is

e.g. row id's may change as the db is moved, and these may be used in URIs

<Souri> rowids may change due to data movement within the same database

souri: not clear why we would do this, row ids can change even without moving the database

ashok: let's wait to try and resolve this issue
... would you like a single statement for this

MacTed: could do that, that essentially means it is like a JDBC connection string

<ericP> imo, ODBC URLs are hilariously varied

<ericP> i often move stuff from server to server without changing primary keys

<boris> also, the connection details can be optional

<boris> ?

<ericP> in fact, all of the databases i download (1/6 months) have pks which have been pre-filled

<Ashok> ACTION Ted to create proposal for issue 37

<trackbot> Created ACTION-138 - Create proposal for issue 37 [on Ted Thibodeau - due 2011-07-05].

This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found ScribeNick: dmcneil
Inferring Scribes: dmcneil

WARNING: No "Topic:" lines found.

Default Present: dmcneil, Ashok_Malhotra, MacTed, +1.562.686.aaaa, Souri, Seema, +539149aabb, boris, ericP, juansequeda, +1.314.288.aacc
Present: Souri David Ted Juan Marcelo Ashok Seema Boris Eric
Regrets: Michael Richard Ivan
Got date from IRC log name: 28 Jun 2011
Guessing minutes URL: http://www.w3.org/2011/06/28-rdb2rdf-minutes.html
People with action items: souri

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report