RDB2RDF Working Group Teleconference -- 11 Oct 2011

<trackbot> Date: 11 October 2011

<joerg> hi

<joerg> jep, first visit, so calling early

<joerg> ;-)

<scribe> scribenick: ericP

Admin

<mhausenblas> PROPOSAL: Accept the minutes of last meeting http://www.w3.org/2011/10/04-RDB2RDF-minutes.html

APPROVED

Open actions and issues

RESOLVED

RESOLUTION: Accept the minutes of last meeting http://www.w3.org/2011/10/04-RDB2RDF-minutes.html

<mhausenblas> ACTION-159?

<trackbot> ACTION-159 -- Boris Villazón-Terrazas to create RDFa representation of R2RML vocab -- due 2011-09-27 -- OPEN

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/159

<boris> http://www.w3.org/ns/r2rml

<mhausenblas> close ACTION-159

<trackbot> ACTION-159 Create RDFa representation of R2RML vocab closed

<mhausenblas> ISSUE-69?

<trackbot> ISSUE-69 -- datatype sizes -- raised

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/issues/69

ericP: note that providing extension advice encourages conformance of the extensions around e.g. representations of 1 BC

ashok: let's not decide anything until cygri is here

Handover from Soeren to Joerg

joerg: we're working on implementaitons of RDB2RDF
... want to represent our WG and our interests in mapping relational database

Ashok: when will you have an implementation we can look at?

<mhausenblas> Michael: we will take Joerg through the IE procedure as described here: http://www.w3.org/2004/08/invexp

joerg: don't want to say now
... prefer to wait until we have stability

LC comments status

Ashok: this is a formal process
... we much record each question or comment
... we have to reply to it
... and we have to get the commentor's agreement
... last week, we spoke of using a wiki
... this is required 'cause the director will ask us about the resolution of our last call comments

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

mhausenblas: so we agreed that for each LC comment, that an editor update http://www.w3.org/2001/sw/rdb2rdf/wiki/Last_Call and@@or create an issue

<mhausenblas> http://www.w3.org/2011/09/20-RDB2RDF-minutes.html#item03

mhausenblas: current understanding per http://www.w3.org/2011/09/20-RDB2RDF-minutes.html#item03 is that we:
... .. do not use the wiki
... .. address editorial comments directly
... .. create issues for substantive issues
... have editors been keeping up on the comments lists?

juansequeda: i didn't seen any major issues (just editorial advice)

mhausenblas: ivan's comments are marked editorial
... I think there are comments from dominic on R2RML and DM
... [get to work, slackers -- ED]

Implementations

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

mhausenblas: just heard about an implementation from antidot

<mhausenblas> Michael: I added them already

Ashok: ivan asked about what we do about the fact that the implementations page claims that many implementations will be available early 2011

mhausenblas: cygri updated it for D2RQ

ericP: FeDeRate can be checked out, not plans for R2RML now

boris: expect to release morph (R2RML-only) end of 2011

MacTed: Virtuoso relase either out now or in the next couple days supports R2RML and DM

dmcneil: early release of Spyder captured early R2RML
... expect update this year

juansequeda: UltraWrap implements DM, DM++ and older R2RML
... should have an update at the end of the year

mhausenblas: i believe nuno is working on the XSPARQL implementation

ericP: having more than 2 interoperable implementations isn't required for exiting CR, but is good for making an impression

mhausenblas: many language/platform implementations tests the spec well

Test Cases

<boris> http://www.w3.org/2001/sw/rdb2rdf/test-cases/

boris: updated the test cases
... currently 40 R3RML test cases

<boris> 40

<boris> http://dvcs.w3.org/hg/rdb2rdf-tests

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_TC

boris, following cygri's advice, created a test case for each part of the spec

mhausenblas: how will we review test cases?
... do we have a test runner?

boris: we haven't finished this down

<Zakim> ericP, you wanted to talk about EARL

ericP: we asked testers to submit test results in an extension of EARL
... we also provided a test harness for SPARQL protocol

mhausenblas: i think it's useful to use EARL, but not useful to write a harness

Ashok, we have used a less formal for folks to say "i've run this and passed some/most/all tests"

Ashok,:we have used a less formal for folks to say "i've run this and passed some/most/all tests"

<mhausenblas> Michael: Also, we could provide simply instructions how to perform the tests

Ashok: we have used a less formal approach for folks to say "i've run this and passed some/most/all tests"

boris: how easy is it to generate EARL

ericP: if we have on the order of a hundred tests, we probably want to know which tests people passed

mhausenblas: +1 to mapping EARL to HTML
... boris, you can describe the process in the test doc

boris: will integrate into test doc and notify list

<mhausenblas> ACTION: Boris to add section concerning testing in http://www.w3.org/2001/sw/rdb2rdf/test-cases/ (re EARL, input/output) [recorded in http://www.w3.org/2011/10/11-RDB2RDF-minutes.html#action01]

<trackbot> Created ACTION-160 - Add section concerning testing in http://www.w3.org/2001/sw/rdb2rdf/test-cases/ (re EARL, input/output) [on Boris Villazón-Terrazas - due 2011-10-18].

boris: we need to check my new test cases

AOB

ericP: i'm not sure we need a process to approve test cases since we don't intend to include incorrect tests (as we did in the record of SPARQL)

<boris> sorry

<mhausenblas> [meeting adjourned]

<mhausenblas> trackbot, end telecon

- DRAFT -

RDB2RDF Working Group Teleconference

11 Oct 2011

Attendees

Contents