W3C

RDF Data Access WG Teleconference

1 Feb 2005

See also: IRC log

Attendees

Present
AndyS, Yoshio, Kevin, DaveB, JosD, AlbertoR, DanC, SteveH, EricP, PatH
Regrets
KendallC, Howard
Chair
DanC
Scribe
AlbertoR (and DanC)

Contents

 

Convene, take roll, review records and agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0076.html

regrets Kendall Clark

<DanC> Scribe: AlbertoR

<DanC> not done: http://www.w3.org/2001/sw/DataAccess/ftf4.html $Revision: 1.4 $

<DanC> 6 of 25 items summarized

<DanC> next meeting: Tuesday, February 8, 2005

<DanC> partial regrets AndyS

<DanC> next scribe: EricP

EricP will scribe 08/02/2005

<DanC> ACTION: DanC (with EricP) to communicate WBS user experience to systems team.

ftf meetings

<DanC> - ACTION AlbertoR: to propose a f2f meeting in Leiden - DONE http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0072.html

<DaveB-lap> www2005 is 10-14 May

<DanC> "10-14 May 2005 Chiba, Japan

<DanC> WWW2005" -- http://www.w3.org/2001/sw/CG/#future

ok proposed 6th f2f will be in between www2005 and XTech2005 - proposed 19-20 May 2005

<DanC> ACTION: DanC to consider Asemantics hosting proposal w.r.t. WG schedule

SteveH was also willing to offer to host next face-to-face

<DanC> ACTION: SteveH to propose to host in Southampton right after XTech

<DanC> ACTION: PatH to propose to host in Florida

valueTesting, XML Schema datatypes

<DanC> XML Schema Datatypes and the Semantic Web Jeremy Carroll (Monday, 31 January)

<DanC> http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0071.html

<DanC> http://www.w3.org/2001/sw/BestPractices/XSCH/xsch-sw-20050127/

SWBP requested review of the XSCH document

<scribe> ACTION: EricP and PathH will review XSCH draft document

the two reviewers will first report to DAWG - comments will most propably arrive after publication

accessing collections

JosD test-case http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0077.html

<DaveB-lap> for me ,postpone

<SteveH> +1

<AlbertoR> +1

<DanC> postpone, pick an o:item URI, make syntax

<DanC> ^options

<DanC> yoshio, which do you prefer/like? postpone, pick an o:item URI, make syntax

about to postpone accessing-collections:

postpone: 6

<DanC> postpone: 1+1+1+1+1+1 -1; pick uri: 1+1

o:item: 2

<Yoshio> 0

<SteveH> not really realted

DanC: cwm seems/is able to deal with collections nicely - it might be stated into the service description

EricP: mentioned a possible issue related to the interaction of collections (or inferred triples) and result-set format

<AndyS> There is also the &ns:func() mechanism

<SteveH> I'm interested, but still think its early

JosD and PatH (Andy partially?) agree on access-collection importance for this round-

SPARQL Protocol Spec

<DanC> ACTION: KendallC to Announce the protocol spec

<DanC> ACTION: EricP to to inviestigate WSDL/SOAP encodings of SPARQL protocol

Kendall, EricP actions continued

<DanC> ericp: preempted by datatype work

<DanC> ACTION: KendallC to to propose a solution to protocolRootReferent

<DanC> ACTION: DanC to write up "sparqlParam" approach (reply to kendallC)

<DanC> Some protocol & service description issues http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0046.html

<SteveH> no

<DanC> ACTION: SteveH to Write up a service description of features supported for his service

<DanC> ACTION: KendallC to Add section for service descriptions and put graph specifying in it to start it off (some progress, it seems. continues)

<DanC> ACTION: DanC to Write up definitive service concept

<DanC> ACTION: KendallC to to think about predicate to relate html-forms and svc descriptions, and how this goes into the spec

<DanC> ACTION: DanC to discuss accessingCollections in service description context

XML Result set

<DanC> DaveB to discuss options for column ordering and pick one

<SteveH> +1 to query order

<DanC> DaveB: seems "unconstrined" is unpopular, so I'm down to 2

<SteveH> * would be mention order wouldn't it?

<SteveH> unconstrained order is inconvienient for clients

wondering if a var can contain unicode encoded chars e.g. \u2070 ? it seems yes accordingly to current BNF grammar....

<AndyS> What's u2070?

I picked up one from BNF (randomly) - just wondering about Unicode and ordering in the result-set

<SteveH> and little more work for servers

<AndyS> But yes, the var name is currently NCNAMES - any problem there?

<ericP> intellectually, i prefer DanC's. aestheticaqlly, i prefer andy's

<SteveH> SELECT order is very useful in some cases

<AndyS> Not sort order!

<SteveH> especially if you have SELECT expressions!

<AlbertoR> +1 SteveH

<ericP> +1 SteveH

<AndyS> +1 to SELECT order if SELECT ?a ?z ?c

<AndyS> If SELECT * - don't care but can't see value of mention order

<SteveH> AndyS, why not? its helpful sometimes

<SteveH> I think SQL does specify, but havent checked

<AndyS> It's work - assumes (1) you parsed it and (2) you parsed in order :-)

<SteveH> AndyS, yes, but easier for users - ie. predictable webween systems

<DanC> PROPOSED: order in result set is: if SELECT a,b, c then a,b,c; if SELECT * then unconstrained

<SteveH> *between

<SteveH> mildly

<SteveH> I prefer mention order

<DanC> SteveH argues for first mentioned order in SELECT *

<DanC> EricP: mysql seems to use schema order in case of SELECT *

<DanC> SteveH notes that lexicographic order is complicated for expressions

if SELECT *would be sorted in query order - which one? (considering optimizers and arbitrary grap-patterns processing)

<ericP> why is nothing easy?

<DanC> PROPOSED: order in result set is: if SELECT a,b, c then a,b,c; if SELECT * then unconstrained

<DanC> so RESOLVED. SteveH abstaining

<ericP> ok, i guess that was comparitively easy

<DaveB-lap> http://lists.w3.org/Archives/Public/www-rdf-interest/2005Jan/0103.html

<DanC> DONE: ACTION DaveB: to post announcement about XML result set on appropriate mailing lists.

<DanC> ACTION: DanC to ask Liam, via SemWeb CG, which XML groups might be interested in SPARQL result set WD

<SteveH> i've implemeted much of it

<DanC> several seem to have implemented it, or started anyway

<DanC> Turtle Tuples: Turtle-based query result format Arjohn Kampman (Monday, 24 January) http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2005Jan/0019.html

<SteveH> I have a turtleish result for in the current version of my store - its good for low bandwidhth but not hugely important

<SteveH> its very like arjohns

<AndyS> I had a sim format - and dropped it 'cos of maintenance overhead for now

test update

<DaveB-lap> bound tests in http://lists.w3.org/Archives/Public/public-rdf-dawg/2005JanMar/0080.html

<DanC> re ACTION DaveB: to draft tests for BOUND (based on ericp's examples)

<DaveB-lap> AndyS found a bug in the 0080

DaveB: will make a new direcotry in the test cases rep for BOUND

All other actions continued

<DanC> ACTION: SteveH to to propose tests for str() and =~

<DanC> ACTION: SteveH to to revise test manifest w.r.t. "background" and named graphs

<DanC> ACTION: DaveB to to propose source test to approve

<DanC> ACTION: AndyS to to propose test case re optionals and ordering

SPARQL QL (rq23) update

<DanC> http://www.w3.org/2001/sw/DataAccess/rq23/ $Revision: 1.4 $ of $Date: 2005/02/01 15:35:02 $

<DanC> ACTION: EricP to to update rq23 w.r.t. str()

<DanC> ACTION: EricP to to update rq23 w.r.t. BOUND/UNSAID (some progress; UNSAID removed)

<DanC> EricP seems to have new info re funcs/ops. will send.

AndyS:anyURI is about a literal with a URI datatype

<DanC> ACTION: AndyS to to revise "unnamed" terminology (continues after all)

<DanC> ADJOURN at 15:35 UTC.

Summary of Action Items

[NEW] ACTION: AndyS to to propose test case re optionals and
... ordering
[NEW] ACTION: AndyS to to revise "unnamed" terminology (continues
... after all)
[NEW] ACTION: DanC to ask Liam, via SemWeb CG, which XML groups
... might be interested in SPARQL result set WD
[NEW] ACTION: DanC to consider Asemantics hosting proposal w.r.t.
... WG schedule
[NEW] ACTION: DanC to discuss accessingCollections in service
... description context
[NEW] ACTION: DanC to to (with EricP) communicate WBS user
... experience to systems team.
[NEW] ACTION: DanC to write up "sparqlParam" approach (reply to
... kendallC)
[NEW] ACTION: DanC to Write up definitive service concept
[NEW] ACTION: DaveB to to propose source test to approve
[NEW] ACTION: EricP and PathH will review XSCH draft document
[NEW] ACTION: EricP to to inviestigate WSDL/SOAP encodings of
... SPARQL protocol
[NEW] ACTION: EricP to to update rq23 w.r.t. BOUND/UNSAID (some
... progress; UNSAID removed)
[NEW] ACTION: EricP to to update rq23 w.r.t. str()
[NEW] ACTION: KendallC to Add section for service descriptions and
... put graph specifying in it to start it off (some progress, it
... seems. continues)
[NEW] ACTION: KendallC to Announce the protocol spec
[NEW] ACTION: KendallC to to propose a solution to
... protocolRootReferent
[NEW] ACTION: KendallC to to think about predicate to relate
... html-forms and svc descriptions, and how this goes into the
... spec
[NEW] ACTION: PatH to propose to host in Florida
[NEW] ACTION: SteveH to propose to host in Southampton right after
... XTech
[NEW] ACTION: SteveH to to propose tests for str() and =~
[NEW] ACTION: SteveH to to revise test manifest w.r.t. "background"
... and named graphs
[NEW] ACTION: SteveH to Write up a service description of features
... supported for his service
[NEW] ACTION: SteveH will make another offer for next face-to-face
 
[End of minutes]
Minutes formatted by David Booth's scribe.perl version 1.109 (CVS log)
$Date: 2005/02/01 15:35:02 $