<trackbot> Date: 14 December 2010

<mhausenblas> betehess, are you gonna join?

<mhausenblas> ericP?

:)

<mhausenblas> scribenick: nunolopes

<mhausenblas> hey rrr

Admin

<mhausenblas> PROPOSAL: Accept the minutes of last meeting, see http://www.w3.org/2010/12/07-rdb2rdf-minutes.html

<boris> +1

<hhalpin> +1

<betehess> +1

RESOLUTION: WG Accepted minutes from last week

<mhausenblas> ACTION-79?

<trackbot> ACTION-79 -- Michael Hausenblas to prepare a SemTech Tutorial -- due 2010-12-14 -- OPEN

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/79

Ashok: I suggest we start with topic 5 from the agenda

mhausenblas: ok

R2ML mapping to ontologies

<mhausenblas> our charter says:

<mhausenblas> [[

<mhausenblas> The mapping language specification SHOULD include guidance with regard to

<mhausenblas> mapping relational data to a subset of OWL such as OWL QL or OWL RL."

<mhausenblas> ]]

Ashok: this is one of the documents part of the charter that we have to publish

<mhausenblas> Michael: see http://www.w3.org/2009/08/rdb2rdf-charter

<hhalpin> maybe make it part of the R2RML doc or an appendix?

… we should get someone to start thinking about it

… I was hoping Percy could do it

<hhalpin> juan claims that he and marcelo were working on this, but neither are here.

mhausenblas: what do you except here, guidance?

Ashok: in charther, in the deliverables section it mentions RDF schema and OWL

hhalpin: the requirement would be that the schema would be mapped to RDFS or OWL

<betehess> [[ /me also reads: The mapping language SHOULD use W3C RIF whenever a rule engine is needed in the mapping language. ]]

… we can think of this has being part of the language

… or produce a seperate document

Ashok: you had talked about doing this with direct mapping, how would this work?

<mhausenblas> http://www.w3.org/2001/sw/rdb2rdf/r2rml/

<mhausenblas> @@Issue: As currently defined, the output of an R2RML mapping is an RDF dataset that contains triples reflecting the records in the database. The output is thus on the "instance level". A feature under discussion is the additional generation of an RDF Schema or OWL Ontology that reflects the constraints and valid inferences over the classes and properties in the generated dataset. This output would be on the "terminology level".

mhausenblas: in the current version of the editors draft, the introduction says :

… could we turn this note into and ISSUE

<hhalpin> maybe send this in e-mail out over the list as well to catch attention of eric, marcelo, and juan?

… and from that derive what needs to be done

<ivan> +1 to have these properly on the issue list

Souri: I was confused about the requirement

<hhalpin> ahhh

… I thought we had to create an ontology for the R2RML vocabulary itself

<hhalpin> that could be useful as well.

… I have a draft of this

… but it seems the requirement is not that

mhausenblas: now I just want to make sure the process is properly recorded

<betehess> +1 to have distinct issues

Souri: what we are supposed to generate is from the relational data

… map to OWL QL, subset which captures relational databases

<mhausenblas> PROPOSAL: Souri adds issues to http://www.w3.org/2001/sw/rdb2rdf/track/issues/ , one for Ashok's concern and one re each of the open, inline-Issues in http://www.w3.org/2001/sw/rdb2rdf/r2rml/

… the inferences would then be limited by OWL QL

<Zakim> betehess, you wanted to say a word about priorities for Direct Mapping: first the data, then OWL

betehess: for the direct mapping, we can think about generating extra triples

ivan: in case I have an existing ontology that is not OWL QL?

<Zakim> mhausenblas, you wanted to talk about process

Souri: I was not thinking of an existing ontology, we may want to combine the ontologies yes

mhausenblas: I have to insist on the fact that we want to have ISSUES

… on the tracker

… to assign actions to it

<mhausenblas> ACTION: Souri to adds issues to http://www.w3.org/2001/sw/rdb2rdf/track/issues/ , one for Ashok's concern and one re each of the open, inline-Issues in http://www.w3.org/2001/sw/rdb2rdf/r2rml/ [recorded in http://www.w3.org/2010/12/14-rdb2rdf-minutes.html#action01]

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

<Souri> sdas2

<Seema> use sads2

<mhausenblas> ACTION: sdas2 to adds issues to http://www.w3.org/2001/sw/rdb2rdf/track/issues/ , one for Ashok's concern and one re each of the open, inline-Issues in http://www.w3.org/2001/sw/rdb2rdf/r2rml/ [recorded in http://www.w3.org/2010/12/14-rdb2rdf-minutes.html#action02]

<trackbot> Created ACTION-88 - Adds issues to http://www.w3.org/2001/sw/rdb2rdf/track/issues/ , one for Ashok's concern and one re each of the open, inline-Issues in http://www.w3.org/2001/sw/rdb2rdf/r2rml/ [on Souripriya Das - due 2010-12-21].

… please make sure to flag the issue with the proper product

R2RML linked data aspects

mhausenblas: the charter is even stronger in this aspect

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

… devise a mechanism for well known entites to have URIs

… I can take an action to create an issue for that

hhalpin: just to remind that angela was interested in this as well

… we can think of using some API calls to solve this

mhausenblas: yes, good, something like opensearch

… but again we are only defining the problem now

<hhalpin> My time is quite full, but I can e-mail the idea out to the list.

<hhalpin> or happy to review on weekends!

<Ashok> There is also the OKKAM work

<Ashok> Harry, please copy Angela on the note ... maybe we can get her interested again!

<hhalpin> ok

mhausenblas: OKKAM was an EC project and are currently working on future steps

<hhalpin> I will send out a quick note

<mhausenblas> ACTION: Hausenblas to create a Linked Data issue and draft proposal to address [recorded in http://www.w3.org/2010/12/14-rdb2rdf-minutes.html#action03]

<trackbot> Created ACTION-89 - Create a Linked Data issue and draft proposal to address [on Michael Hausenblas - due 2010-12-21].

… we should take care not to rely on tech we are working on

mhausenblas: would this be R2RML or direct mapping?

juansequeda: Ashok: R2RML

<Zakim> betehess, you wanted to say that Eric and I don't plan to solve the Linked Data aspect within the Direct Mapping. we want to chain the Direct Mapping with other technologies (RIF,

betehess: just to talk about the aproach eric and I are working on

<hhalpin> sounds good

… by using SPARQL constructs we can rewrite the query directly

<hhalpin> I think Alex just said that R2RML could handle that visa SPARQL or RIF

<ivan> +1 to betehess, b.t.w.

<hhalpin> but R2RML may want an easier plug-in mechanism

mhausenblas: we can start the issue under R2RML and if needed we change it to direct mapping

<betehess> hhalpin, I don't know for R2RML, but this is what we do for the Direct Mapping :-)

dissemination

<mhausenblas> ACTION-79?

<trackbot> ACTION-79 -- Michael Hausenblas to prepare a SemTech Tutorial -- due 2010-12-14 -- OPEN

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/79

mhausenblas: juansequeda created the wiki page regarding this

<mhausenblas> http://www.w3.org/2001/sw/rdb2rdf/wiki/RDB2RDF_Publicity

… there are venues

… the other issue was to come up with a concrete proposal for semtech

… both ivan and juan are in the commitee

… so maybe it's better if I formalise the submition to avoid conflicts

juansequeda: ok

mhausenblas: we should start working on the proposal and circulate it among the WG

… if there is no telco on the 4th we won't have a change to talk about it before the deadline

… we can put something on the wiki and ask fro input from the WG

juansequeda: I can start doing this on the wiki

mhausenblas: let's discuss the options of submition with the WG

juansequeda: we should have a 1h conference session

<mhausenblas> Michael: Full Conference Sessions (50min) OR Half-Day Tutorials (3h)

<boris> or both?

<Souri> 1.5 hr extended session ?

… other thing is if we should have a tutorial as weel

<betehess> a tutorial would mean you need to have a working prototype

<mhausenblas> http://semtech2011.semanticweb.com/cfp.cfm?pgid=40

mhausenblas: I propose full conference session of half day tutorial

betehess: for the tutorial do we need a prototype?

<betehess> +1 to Ivan

ivan: our work is probably not suitable for a tutorial

<juansequeda> +1 to Ivan

… tutorial would mean that language would be close to finished

<Ashok> So, a 50 minute or 90 minute session?

… tutorial for 2012 is better

mhausenblas: so the proposal would be a 1h sesison?

<mhausenblas> PROPOSAL: We want a Full Conference Session for SemTech

Ashok: can we explain both our documents in 1h?

ivan: I did short presentation on both in about 25 minutes

… not easy but doable

<Souri> +1 to 50-min

<juansequeda> +1

<ivan> +1

<Ashok> +1

<boris> +1

RESOLUTION: WG wants a full conference session for semtech (50m)

<betehess> +1

<mhausenblas> ACTION: Juan to draft a submission for a 50min session for SemTech on Wiki and ping WG mailing list [recorded in http://www.w3.org/2010/12/14-rdb2rdf-minutes.html#action04]

<trackbot> Created ACTION-90 - Draft a submission for a 50min session for SemTech on Wiki and ping WG mailing list [on Juan Sequeda - due 2010-12-21].

<ivan> i am

<ivan> i plan to

juansequeda: who knows if they will be at semtech?

mhausenblas: when you send out the draft maybe ask who is planning on participating

<hhalpin> would be a good place to do a panel, but not planning on going

<ivan> not planning to go:-(

juansequeda: there is also the possibility of doing a panel at ESWC

<hhalpin> I imagine it depends on paper acceptances

mhausenblas: is somebody is going

… let's see if we have participants

… please also ask in an email

juansequeda: I'm still going to update the venue list on the wiki

… eventually we want to have a tutorial , deadline for ESWC is may

review R2RML testcases

<mhausenblas> ACTION-81?

<trackbot> ACTION-81 -- Juan Sequeda to and Eric to draft Direct Mapping TC -- due 2010-12-14 -- OPEN

<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/81

<mhausenblas> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1

boris: the new testcases are still not in the wiki

<betehess> /me is eager to have a machine readable test suite

<betehess> our prototype is ready to work with that

… according to eric's proposal to reorganine the document is in the wiki

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#DB0

… so far we have one test case on hold

… we are wainting for the direct mapping

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#DB1

… one is accepted

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#R2RMLTC0001b

… next testcase

… tests 1 colimn mapping with no primary keyt

… so we have a bnode for the subject

… and map one column to a property

<boris> PROPOSAL: Accept http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#R2RMLTC0001b

<mhausenblas> +1

RESOLUTION: Accepted http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#R2RMLTC0001b

<ivan> looks o.k to me...

<Souri> +1 (it took me a little longer)

… next testcase

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#DB2

<hhalpin> I am not sure if the test cases need to be gone through one by one....I assume bugs will come out when implementations try to conform to them.

<boris> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#R2RMLTC0002a

… here we have a table with 2 coluns and no primary key

… the uri is generated by concatenating the two values

<betehess> should we accept if there is no Direct Mapping yet?

mhausenblas: the syntax and testcases mutually influence each other

<boris> for that database, there will be a default mapping test case http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#Direct_Graph_1

… and these will probably form the basis for implementations

boris: for each database we will have a default mapping testcase

… and one for the R2RML syntax

Souri: A comment on 2a

… nevermind

boris: any qestions?

<Zakim> ivan, you wanted to comment on the overall methodology

ivan: not on the specific testcase but the methodology

… 2 issues, the way we are doing this it will not scale

<scribe> … new testcases should be sent by email

<Souri> +1 to Ivan's proposal

… and unless there is a negative review within a week this will become a testcase

… the other problem is that I'm not sure how we want to use the testcases

<hhalpin> +1

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

… will the implementor have to cut-and-past the testcases

mhausenblas: we can change the methodology

… in the RDFa taskforce we initially reviewed the testcases

… and for new ones we checked it

…for the other question we will have a manifest document with the testcases in RDFa

<betehess> we should use this format even for no yet reviewed tests

<betehess> not wiki

… automated testing using for instance mysql

ivan: should we put a reference for the RDFa

<mhausenblas> see http://rdfa.digitalbazaar.com/test-suite/

… the editor's draft should behave as a manifest file

… with the testcases as RDFa

… and we can even use automated reporting tools

<betehess> mhausenblas, please ask for votes for a PROPOSAL about the test suite: even not reviewed tests should follow your RDFa approach

betehess: we should not use the wiki, but have 2 pages

… one for approved testcases and another for non aproved

<boris> 2 pages?

<boris> not wiki pages?

mhausenblas: you mean to have 2 pages in the wiki

betehess: no, on a differnet URL

… but have it available even if the WG has not accepted it

ivan: for the RDFa testsuite we can choose to see only approved testcases or all of them

mhausenblas: for instance using a SPARQL query

<mhausenblas> http://www.w3.org/2006/03/test-description

…. to select which testcases to see

… so we don't necessarily need 2 pages

mhausenblas: do we want to change the methodology?

… for the test case approval

ivan: when boris has a new test he would send out an email

mhausenblas: how about the exiting ones

?

ivan: for those I don't know

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

boris: maybe I can send an email with the status of each

mhausenblas: we can assume the non accepted ones are new testcases

<mhausenblas> ACTION: Hausenblas to move accepted TC from wiki to http://www.w3.org/2001/sw/rdb2rdf/test-cases/ [recorded in http://www.w3.org/2010/12/14-rdb2rdf-minutes.html#action05]

<trackbot> Created ACTION-91 - Move accepted TC from wiki to http://www.w3.org/2001/sw/rdb2rdf/test-cases/ [on Michael Hausenblas - due 2010-12-21].

<hhalpin> +1

… and follow ivan's suggestion

AOB

mhausenblas: I would suggest to resume on the 11th of Jan

… not sure if I will be around

<juansequeda> +1

<boris> +1

<boris> actually is 12

mhausenblas: we'll resume on 11/1/11

<ivan> have a good rest everyone!

bye

<betehess> bye

<mhausenblas> [adjourned]

<mhausenblas> trackbot, end telecon

This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/also reads [[ The mapping language SHOULD use W3C RIF whenever a rule engine is needed in the mapping language/also reads: The mapping language SHOULD use W3C RIF whenever a rule engine is needed in the mapping language/
Succeeded: s/SPARLQ/SPARQL/
Found ScribeNick: nunolopes
Inferring Scribes: nunolopes
Default Present: boris, mhausenblas, Ivan, hhalpin, Ashok_Malhotra, nunolopes, Alexandre, Souri, Seema, +575737aaaa, juansequeda
Present: boris mhausenblas Ivan hhalpin Ashok_Malhotra nunolopes Alexandre Souri Seema +575737aaaa juansequeda
Regrets: Ted Marcelo Percy
Agenda: http://lists.w3.org/Archives/Public/public-rdb2rdf-wg/2010Dec/0019.html
Found Date: 14 Dec 2010
Guessing minutes URL: http://www.w3.org/2010/12/14-rdb2rdf-minutes.html
People with action items: hausenblas juan sdas2 souri

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