W3C

RDB2RDF Weekly Telecon

09 Feb 2010

Attendees

Present
Souri, juansequeda, Ashok_Malhotra, hhalpin, MacTed, mhausenblas, +039046188aaaa, nunolopes, Orri, EricP
Regrets
Soren, Ben
Chair
Ahmed
Scribe
nunolopes

Contents


<ericP> ahh, i'm supposed to be elsewhere now

<mhausenblas> ericP ping

<mhausenblas> I'm at risk - need to sort out two things urgently and dunno how long it takes. Richard and Nuno will call in via my phone in my office, hope to join in later on ... sorry for late notice

<MacTed> hm. lacking a zakim is going to make things interesting.

<MacTed> oh good, invitations work

trackbot, start telecon

<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)

trackbot, start telecon

<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, help

<trackbot> See http://www.w3.org/2005/06/tracker/irc for help

<MacTed> trackbot, init

<MacTed> trackbot, reload

<MacTed> trackbot, start telecon

<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, status

<trackbot> This channel is not configured

<MacTed> not good.

<scribe> scribenick: nunolopes

<MacTed> sdas2 - IRC nickname changes with `/nick Souri` ... that's outside Zakim's control

<MacTed> hhalpin - if you know the mysteries of trackbot, that may be good.

roll call

<scribe> scribenick: nunolopes

<hhalpin> I am here...

Minutes from last week

Ahmed: there is an error in the minutes

<Souri> could you please post the URL for the minutes? Thx.

<juansequeda> http://www.w3.org/2010/02/02-RDB2RDF-minutes.html

http://www.w3.org/2010/02/02-RDB2RDF-minutes.html, I think

Continued discussion about different approaches

<Souri> +q

Souri: what is not needed in the schema is a view creation

Ahmed: there are enviroments when we cannot change the schema

<cygri> sorry i'm late -- i'm now here with mhausenblas/nunolopes

… we need to support both

<hhalpin> +1 Ahmed

<Souri> Our approach mentions only the SQLdefString (i.e., the query) --- there is no *need* to create a DB view

Orri: we might want to materialize views in some cases

Ahmed: we need to consider both approaches

… although no change of the database schema seems more used

Souri: two groups can have 2 working groups on the wiki

… discussion over email

… if more interaction is needed we can do it during the telco

hhalpin: telco should be held even if everyone agrees

<Zakim> ericP, you wanted to suggest that we don't need to know the full self-definition of the groups to have them work on e.g. use cases and therefor expressivity and to suggest both

<MacTed> trackbot, status

<trackbot> This channel is not configured

ericP: if we don't have anything to talk we close early (should be rare)

<ericP> http://esw.w3.org/topic/HCLSIG/RDB2RDF_Use_Cases

… for use case documents, both groups should contribute

<hhalpin> well, it would be nice

… the wiki version documents the SPARQL query and the SQL mapping

Ahmed: can we have a model where someone makes a big change to the wiki the group gets notified?

<Souri> +q

hhalpin: we can set that up

… i'll put the instructions on the wiki

<Souri> -q

… it's part of the wiki, you can put yourself as a member who wants to get notified

Ahmed: for both groups I want someone to lead

… I don't want to lead any group

ericP: I can step up for the mapping approach

Souri: I can volunteer for the SQL approach

<hhalpin> we need better names!

Ahmed: do you mean the schema change approach?

Orri: you can have RDF and SQL approach where neither requires changes to the schema

<Souri> SQL-based approach and RDF-based approach

ericP: we should choose 2 labels for the groups

Orri: SQL style like oracle

… RDF style like D2RQ

Ahmed: are these 2 cases independant of changing the schema?

Orri: SQL style does not presume we can change the schema

… changing the schema is a matter of creating views

… this can be circumvented by using inline queries

… you can do this if you have permissions

Ahmed: what about issues with mysql?

cygri: the RDF style seems close to what we do in D2RQ

<hhalpin> ACTION: hhalpin to update website

<trackbot> Created ACTION-24 - Update website [on Harry Halpin - due 2010-02-16].

Ahmed: can the website be updated with the names of the approaches?

<hhalpin> +1 souri

Souri: SQL approach is secondary if you change the schema

… I can volunteer for the SQL based approach

Ahmed: ericP for the RDF approach and Souri for the SQL approach

ericP: what document are we preparing?

<Souri> +q

… we should figure out what the use cases are

juansequeda: given our current approaches can we define that D2R is the RDF approach and openlink the SQL?

Ahmed: that's a strong assumption to make

Orri: SQL style would be triplify

… RDF, D2RQ and Virtuoso

Ahmed: there are at least 2 approaches for each adn we should discuss them

… regarding the document, conflicts will arise

… how long are we going to wait for the use cases document?

… I would like to move

Souri: use cases documents are important

<ericP> i don't think we're going to satify the community without a WG use case document

… for which ever approach we use the mapping

… looking at the examples we see which is the schema, SPARQL query

… both groups need to agree what the RDF schema data should be for the query designer

<hhalpin> we have till April in charter for use-case note.

Ahmed: how long to define the schema and use-cases?

Souri: the mapping of the relational data to RDF, how would it look like if it were materialised?

Ahmed: are we going to define the view?

Orri: we can give examples for the mappings, views and data

ericP: what the world needs is a mapping language

… as long as the system gives the appearence of the virtual RDF graph doesn't matter how you do it

Ahmed: I would like to see examples of how that would work

Orri: makes sense to take examples from SQL like some benchmark suites

… and show how they would look in SPARQL

hhalpin: we need to have the use cases in April

… draft of it by march

… we can devote one telco to use cases document

… invite someone from HCLS group

ericP: i'm doing both

<hhalpin> Its due end of end of April...

Ahmed: ericP can you edit the use cases?

<hhalpin> the use case document that is.

Ahmed: any volunteers to edit the use cases?

Souri: if ericP can help me to get started I can do it

ericP: ok

Ahmed: we have editors for the use-cases and leads for the two teams

<hhalpin> ACTION: ericP and Souri to own the use-case documents

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

<hhalpin> ACTION: Eric and Souri to own the use-case documents

<trackbot> Created ACTION-25 - And Souri to own the use-case documents [on Eric Prud'hommeaux - due 2010-02-16].

Ahmed: anything else?

hhalpin: the draft of the mapping should be out in june

… have drafts by march

… face to face at SemTech

Ahmed: Was this face to face already decided?
... can we make it official?

<hhalpin> PROPOSED: Face-to-face at SemTech 2009?

<hhalpin> going once...going twice...

<hhalpin> RESOLVED: Face-to-face at SemTech 2009

ericP: when exactly the face to face, before, during or after Semtech?

<hhalpin> We could do Sunday...

… 21 (Monday) to 26 of june

Ahmed: we can meet sunday

ericP: do we want a 2 day face to face?

<hhalpin> we could so saturday->sunday (I'm not a father though..)

ericP: Saturday and Sunday before semtech?

Ahmed: I can make it on Saturdays

ericP: we can try Sunday and Monday

hhalpin: we can do one day after semtech since we can get some feedback during semtech

ericP: we can meet two mornings during semtech

<juansequeda> http://www.w3.org/2009/12/rdf-ws/ RDF workshop (but apparently the date on the website is wrong)

ericP: would anyone go for the face to face without going to semtech? if not we can do one day before and one day during semtech

Ahmed: on monday usually there are paid tutorials

… people can be reluctant to leave

ericP: how about Friday? towards the end of the conference?

Ahmed: I agree

Souri: I agree

<hhalpin> +1

ericP: 20th all day, 25th half day

… will send mail to the list

Souri: are we doing any panel at semtech?

ericP: we submitted but are waiting for confirmation

Ahmed: can someone find out if there will be a panel?

<hhalpin> trackbot, draft minutes

<trackbot> Sorry, hhalpin, I don't understand 'trackbot, draft minutes'. Please refer to http://www.w3.org/2005/06/tracker/irc for help

<hhalpin> trackbot,start meeting

<hhalpin> ?

<hhalpin> hmm

<hhalpin> trackbot is off kilter

no minutes?

<hhalpin> well, for some reason the auto-logging didn't start.

<hhalpin> we can do it manually, it's just a pain.

:(

I have no idea how to do it

<hhalpin> dont worry nuno I'll fix this

Summary of Action Items

[NEW] ACTION: Eric and Souri to own the use-case documents
[NEW] ACTION: ericP and Souri to own the use-case documents
[NEW] ACTION: hhalpin to update website
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/02/23 17:30:38 $