W3C

RDB2RDF Working Group Teleconference

07 Sep 2010

Agenda

See also: IRC log

Attendees

Present
Eric, Nuno, Richard, Souri, Ted, Michael, Juan, Dan, Seema
Regrets
Chair
Michael
Scribe
juansequeda

Contents


<trackbot> Date: 07 September 2010

<MacTed> hrm.

<MacTed> broken bot.

<MacTed> bah

<cygri> hm, we can try this:

<cygri> nunolopes, who's here?

<cygri> nunolopes, i'm with mhausenblas

<Ashok> s/om/on/

<mhausenblas> seems we're having issues with Zakim

<nunolopes> cygri: this bot is even more broken then the other

<ericP> aha!

<mhausenblas> ericP is very welcome

<mhausenblas> scribenick: juansequeda

Admin

<mhausenblas> PROPOSAL: Accept the minutes of last meeting, see

<mhausenblas> http://www.w3.org/2010/08/31-rdb2rdf-minutes.html

<cygri> the ambulance is for zakim

+1 for minutes

RESOLUTION: WG has accepted the minutes of past meeting

Default Mapping

mhausenblas: there has been discussion on this in the last few hours
... will send the agenda earlier because this triggers discussion

juan: suggest to move on to topic 3 (syntax) and then go back to default mapping later

Syntax

<mhausenblas> http://www.w3.org/2002/09/wbs/43889/RDB2RDF-syntax/results

mhausenblas: comments on the syntax?
... questions?

cygri: so far, the answers are from the people who are deeply involved in the discussion
... it would be interesting to get popular vote from the people who haven't thought about these issues
... how to do this?
... would it make sense to make this a public poll?

mhausenblas: I was expecting feedback from the group on this
... probably people involved in RDB2RDF are better informed to make this decision
... but opening to the public is an option

ericP: I think that is a good idea. interested to see what consumers would like to type

mhausenblas: should we have a separate poll? or open this one to the public

ericP: need to get permission
... from everybody who has already submitted

<ericP> +1

<Ashok> By "public" do you mean W3C members or man-in-the-street?

cygri: I will volunteer to make the custom syntax example can be the same as the others

mhausenblas: wrt Ashok's question... it is open to either

<Souri> Seema and I have added a new (revision 2) page for XML-schema based example: http://www.w3.org/2001/sw/rdb2rdf/wiki/Example_of_SQL-based_RDB2RDF_Mapping:_Revision_2

Souri: in the DB community, I don't know how much is connected to this group. The feedback that we will get from the community is mostly going to be from the RDF side

mhausenblas: this is the opportunity to do PR. We can approach people from Semtech

Souri: the DB people may not care about RDF
... the db people are not enthusiastic in RDF, but will be players.
... the feedback will come from RDF people

cygri: the people who want to do this are the RDF people and we hope that the DB people will come later
... if we believe something is better for DB people, then it is just speculation
... it is reasonable to design for people that we know who will use it

Souri: I agree that we need to have a RDF syntax.
... we should be open to the XML syntax
... with an open poll, we will probably get feedback about RDF and not about XML

cygri: from the poll, we can see that the RDF syntax should be done
... I don't think anybody will object for the RDF syntax
... some people say that the custom and xml is not a good idea
... We could start with the RDF syntax for the FWPD
... expressing the same thing in another syntax is not a huge problem when we already have nailed down what we want in the first syntax

Souri: yes, let's make an equivalent XML syntax on the way

<Zakim> ericP, you wanted to say that we can ask for explicit feedback in FPWD

mhausenblas: it seems that we have a new proposal of going forth with the RDF syntax with a synchronized XML syntax

ericP: if we show in the doc both syntaxes, we can have the question in the doc if people like this

<mhausenblas> PROPOSAL: use RDF/Turtle in FPWD + keep XML syntax in sync and ask for feedback in FPWD

<ericP> second

<cygri> +1

mhausenblas, everybody agrees?

+1

RESOLUTION: WG has decided to use RDF/Turtle in FPWD + keep XML syntax in sync and ask for feedback in FPWD

Souri: for the RDF syntax, we need to work hard so it can stay simple and it corresponds to each of the things in XML
... there should be a 1:1 correspondence

cygri: in the RDF draft, I took the XML example and tried to write it in RDF. There are a few things that I didn't know how to write them
... so I know there are some things that don't match
... we need to start writing them down, and see what them mean and how they work

Souri: if I can look at the RDF, how can I translate it to XML and vice-versa

FPWD

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

mhausenblas: I would suggest to work directly on the document and not on the wiki
... who volunteers to edit this?
... we need people in the next 2 weeks to start writing this editor draft

<Souri> I don't mind editing with Richard (and if Seema agrees)

cygri: I can

<Seema> I can help out too

<talking logistics of cvs >

<mhausenblas> Michael: editors of the FPWD are Souri, Seema and Richard

mhausenblas: what is the title of that document?

<Souri> Language for Mapping Relational Data to RDF ?

Relational Database to RDF Mapping Language

<mhausenblas> PROPOSAL: R2RML - Relational Database to RDF Mapping Language

+1

<discussion about title>

cygri: is R2RML the fixed name? or can it be changed?

mhausenblas: yes it can be changed

ericP: ML usually means Markup Language

<ericP> PROPOSAL: Relational Database to RDF Mapping Language

ericP: don't give a name for the language... yet

mhausenblas: but we used R2RML for the Use case doc

ericP: so are we committed?

<mhausenblas> +1 ericP's proposal

macted: we need to have a place holder

mhausenblas: It doesn't need to be in the title, but we can still use R2RML as a placeholder

ericP: if it is not used in the title, then we can change the name of the language later

<MacTed> +1 title change

RESOLUTION: WG decided that the FPWD will have the title "Relational Database to RDF Mapping Language"

mhausenblas: how will the editors plan to incorporate the semantics

<mhausenblas> PROPOSAL: having the semantics as a separate document

juansequeda: recommend to have different document for the direct mapping and semantics

cygri, didn't catch what you said

cygri: in two weeks we can have something worth discussing

<cygri> juansequeda: i suggested to first create a skeleton of the document addressing the syntax, and when that is done discuss about how to fit the semantics in (same or separate doc)

<mhausenblas> Michael: Ashok are you available for chairing next week?

cygri, thanks and +1

<Souri> I agree, 2 weeks is more realistic (we'll be out Sep 19-23, and at risk for sep 22)

<cygri> there's a language "R2R" already in the RDF world: http://www4.wiwiss.fu-berlin.de/bizer/r2r/

<Ashok> Yes, I should be able to chair next week

<mhausenblas> great, thanks!

ericP: could we get the direct mapping doc out quickly, because it has been 3 months since our last publication

<mhausenblas> focusing on direct mapping next week, ok?

mhausenblas: let's discuss the direct mapping next week

Souri: I won't be in 2 weeks
... it's fine if Richard presents

AOB

mhausenblas: good meeting. next week Ashok will chair and we will talk about the direct mapping
... sooner or later we should have a f2f meeting

<Ashok> I suggest after we have a FPWD

mhausenblas: it might make sense that once we have the first draft, we can have a f2f so we can keep editing

<mhausenblas> [adjourned]

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/09/07 16:57:07 $