See also: IRC log
<trackbot> Date: 23 February 2010
<Ahmed> Hello...
<Ashok> Hello Ahmed ... I will dial in soon
<soeren> +soeren
<mhausenblas> iv_an_ru, it would be you scribing today, is this ok with you?
<mhausenblas> looking at http://www.w3.org/2001/sw/rdb2rdf/wiki/ScribeList
<mhausenblas> cygri is with mhausenblas
<mhausenblas> http://www.w3.org/2010/02/16-rdb2rdf-minutes.html
<mhausenblas> RESOLUTION: accept the minutes from last telecon http://www.w3.org/2010/02/16-rdb2rdf-minutes.html
<mhausenblas> ACTION-19?
<trackbot> ACTION-19 -- Harry Halpin to put Souri's draft into a WBS poll and notify the WG -- due 2010-01-19 -- OPEN
<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/19
<mhausenblas> ACTION-24?
<trackbot> ACTION-24 -- Harry Halpin to update website -- due 2010-02-16 -- OPEN
<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/24
<mhausenblas> ACTION-25?
<trackbot> ACTION-25 -- Eric Prud'hommeaux to and Souri to own the use-case documents -- due 2010-02-16 -- OPEN
<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/25
<hhalpin> website updated
<mhausenblas> close ACTION-25
<trackbot> ACTION-25 And Souri to own the use-case documents closed
<mhausenblas> close ACTION-24
<trackbot> ACTION-24 Update website closed
<mhausenblas> ACTION-26?
<trackbot> ACTION-26 -- Michael Hausenblas to create F2F Wiki page -- due 2010-02-23 -- OPEN
<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/26
<mhausenblas> ACTION-27?
<trackbot> ACTION-27 -- Michael Hausenblas to invite Olaf some time in the future (will be decided by group) -- due 2010-02-23 -- OPEN
<trackbot> http://www.w3.org/2001/sw/rdb2rdf/track/actions/27
<mhausenblas> actually juansequeda would be next on the scribe list, true ;)
ok I guess I can scribe
<mhausenblas> cool, thanks juansequeda
<mhausenblas> scribenick: juansequeda
<soeren> http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_SQL_View_Annotation
Soeren: will put a syntax document on the wiki
'
scribe: trying to implement this
syntax.
... implement it for Triplify tool and also in Java
... PHP version is already running with basic graph pattern,
but it needs to be extend to work with optionals and
filters
... hope to have an implementation to show if this approach
works
Ahmed: Are you trying to do both PHP and Java?
soeren: yes, both. two phd
students are working on PHP and Java. Algorithms are the same.
So it should be very similar
... the PHP is progressing better.
... more resources for the Java implementation because it is
more important for the group
<Zakim> mhausenblas, you wanted to talk about requirements
<mhausenblas> http://www.w3.org/TR/2009/WD-sparql-features-20090702/
<ericP> minutes for 09 Feb
talking about reqs
Ahmed: When should we wrap up use cases
<mhausenblas> +1
<hhalpin> use-cases need to be wrapped up by April!
<hhalpin> which is coming closer than we think :)
Ahmed: proposal: By next week present use cases
mhausenblas: Let's try to have use cases by next week (march 2)
<angela_UNITN> +q
mhausenblas: and we have a buffer week (an extra week) in case needed
<mhausenblas> PROPOSAL: have UC ready till 2 March
+1
Ahmed: any progress done by each team should be reflected on the wiki
<mhausenblas> RESOLUTION: The WG agrees to have the UC ready by 2 March 2010 and derive the requirements from this soon after
ericP: after the usecases are
ready, we can start with the requirements
... then both teams can have their own page for their syntax,
etc
Ahmed: we need to update the page to reflect the new name of the team (SQL and RDF) and make a page for each team.
<Zakim> juansequeda, you wanted to ask about the semantics of the language
juansequeda: soeren, have you guys worked on the semantics of the language?
<cygri> +q
soeren: we have only been working on the syntax. We have not worked on the semantics yet. We should do that later
angela_UNITN: what do you
envision has requirements to write the use cases?
... should they be highlevel or low level use cases?
<mhausenblas> Ahmed: re syntax, focus on ANSI92
Ashok: I would recommend high level usecases
<hhalpin> as regards use-cases, high-level, let's not confuse it with test-cases :)
soeren: every usecase should have a sql schema with it
Ahmed: that would make then the usecases very small because not everybody can share the schema
<hhalpin> as long as we have a english-language description that say, a product manager could understand
<hhalpin> without too much difficulty
soeren: the HCLS usecase has a
schema attached to its usecase
... I will contribute with a web application usecase like
Wordpress
angela_UNITN: I will contribute with a schema and a domain ontology.
cygri: Is there a problem if we create a language that depends on an ISO standard?
<hhalpin> no we can depend on ISO standards
Ahmed: my thought: we need to support ANSI 99
<hhalpin> not a problem from W3C process perspective, as long as the ISO standard is not "in motion"
<hhalpin> but settled
cygri: I haven't come across W3C
recommendation that refers to a standard that is not freely
available
... if you want to implement the language that we are going to
define, you will need access to the SQL standard document
<ericP> most of us don't know what SQL 92 is ('cause it's not free and not on the web)
cygri: however, you need to pay a fee to get this.
ericP: to some degree we can do
what we want, but we may get objections
... it might be a problem
<hhalpin> I have seen W3C recommendations refer to non-free recommendations
<hhalpin> ISO Common Logic comes to mind
<hhalpin> "depend" is a strict word
<mhausenblas> Michael: I think Richard has a very good point. We need to check this
<hhalpin> the point is that someone who reads a W3C recommendation should NOT have to download
<hhalpin> another recommendation that costs money
<hhalpin> to implement this recommendation correctly
<hhalpin> I will check
<mhausenblas> queue
Ahmed: eriP, can you check with Ivan?
<hhalpin> ACTION: hhalpin to check on dependencies with non-free ISO [recorded in http://www.w3.org/2010/02/23-rdb2rdf-minutes.html#action01]
<trackbot> Created ACTION-28 - Check on dependencies with non-free ISO [on Harry Halpin - due 2010-03-02].
Ahmed: we need to provide in our mapping language a way to leverage things that are vendor-specific
<ericP> weird extensions to the language like OFFSET and LIMIT?
Ashok: SQL 99 is a very big
standard and there are things that nobody implements
... many of the W3C specs refer to the UNICODE act. But you
can't get this spec on the web
... so you ought to be ok with SQL
ericP: but we still risk for
people to object
... this is not a policy issue, this is what our community
wants
<cygri> juansequeda, ashok: for the record, you can download the entire unicode spec for free from unicode.org
Ahmed: ericP, can you please ask
Ivan also about SPARQL
... about the max expressivity of SPARQL because we don't want
to have a mapping language that supports more expressivity than
SPARQL
ericP: sometimes to get things done, we omit expressivity that we would like to do just to get things done quickly
<hhalpin> +1 use-case driven approach
<mhausenblas> ACTION: ericP to create Wiki page for RDF team and link from homepage [recorded in http://www.w3.org/2010/02/23-rdb2rdf-minutes.html#action02]
<trackbot> Sorry, couldn't find user - ericP
<hhalpin> hmmmm as regards the semantic mapping, this is where marcelo would be useful
ericP: we have a need to map semantics of SPARQL back to SQL
<hhalpin> as he has done relevant work here
<hhalpin> http://ing.utalca.cl/~jperez/papers/sparql_semantics.pdf
<mhausenblas> ACTION: Eric to create Wiki page for RDF team and link from homepage [recorded in http://www.w3.org/2010/02/23-rdb2rdf-minutes.html#action03]
<trackbot> Created ACTION-29 - Create Wiki page for RDF team and link from homepage [on Eric Prud'hommeaux - due 2010-03-02].
<hhalpin> yes but marcelo is part of the WG
hhalpin: recommend that Marcelo should write the semantics of SPARQL to SQL
<hhalpin> his work has high regard from the formal semanticists
<hhalpin> in Edinburgh at least.
<hhalpin> I also think richard cygniak had work in this area..
<hhalpin> richard?
Semantics preserving SPARQL-to-SQL translation
<hhalpin> well, not only marcelo, but marcelo plus someone else!
<cygri> hhalpin, yes but that was very early in the field, there's much more complete work out now
<hhalpin> ideally, someone else with experience in formal semantics
<hhalpin> double-checking formal semantics is painful :)
Ahmed: the mapping language should not be aware of sql view or rdf view
<hhalpin> well, we'll just see what happens if we can reconcile them, otherwise we chose one.
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/93/92/ Found ScribeNick: juansequeda Inferring Scribes: juansequeda Default Present: +043316876aaaa, whalb, MacTed, +49.322.22.aabb, soeren, Ashok_Malhotra, mhausenblas, +1.512.484.aacc, +039046128aadd, +222220aaee, [IPcaller], EricP, cygri, juansequeda, hhalpin Present: +043316876aaaa whalb MacTed +49.322.22.aabb soeren Ashok_Malhotra mhausenblas +1.512.484.aacc +039046128aadd +222220aaee [IPcaller] EricP cygri juansequeda hhalpin Regrets: Nuno Marcelo Found Date: 23 Feb 2010 Guessing minutes URL: http://www.w3.org/2010/02/23-rdb2rdf-minutes.html People with action items: eric ericp hhalpin[End of scribe.perl diagnostic output]