W3C

XML Schema Patterns for Databinding WG
24 Jan 2006

Agenda

See also: IRC log

Attendees

Present
Paul Biron (HL7)
Jon Calladine (BT)
Paul Downey (BT)
Anthony (Tony) Julian (HL7)
Yves Lafon (W3C)
Ajith Ranabahu (WSO2)
Regrets
Paul Fremantle (WSO2)
Sekhar Vajjhala (Sun)
Chair
PaulD
Scribe
Yves

Contents


Administrivia

Minutes of last week are approved

upcoming f2f meeting, who will attend?

HL7 won't be able to attend

might join remotely

not for the whole day

wso2 will be present

expecting observers, esp from the XML Schema WG

<scribe> ACTION: yves to ensure we have remote access at the TP [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action01]

skeleton documents will appear soon

plan is to publish ASAP, mainly to have the IP issue cleared soon

roadmap, patterns and test suite are the main deliverables for now

ISSUE-3: Mapping Simple Numeric Types with Infinite Value Space

still waiting for text from pauld

some already defined integer types have limits in it

PaulB: xsd:int was targeted as being implementable directly as java ints, for example

ISSUE-6, ISSUE-10: mapping of names

<scribe> -> no progress on text.

PaulB: sent mail on this - in some cases it could be automatable, like kanji

<scribe> ACTION: pauld to contact i18n WG re-mapping of non-US names to programming languages

ISSUE-7: xsi:nil and minOccurs=0

xsi: nil and minOccurs=0

Jon: contacted epicore about the way they handle this currently

they have a matrix describing the behaviour

resolution is to ask epicore to contribute their currently private information to the group

paulb: minoccurs=0 should be used only if the presence is mandatory (in the abstract model)

chair: we have to be careful about submissions made in private emails or conversations. Ideally Erik should submit a proposal to our list, but if it's on soapbuilders, it's possibly in the public domain.

pauld: the main issue being the semantics of combinations of absence of the element and the fact that the element is null

paulb: I don't expect databinding tool to represent nil (or at least some nils/null)

good to have an understanding of what tools are doing right now

<pauld> Jon and I can work on supplying some information as to what happens in practice.

<pauld> ACTION: jcalladi submit propoal based upon Erik's suggestions to soapbuilders (with Erik's permission) [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action02]

<trackbot> Created ACTION-12 - Submit propoal based upon Erik\'s suggestions to soapbuilders (with Erik\'s permission) [on Jonathan Calladine - due 2006-01-31].

ISSUE-12 Identification of Patterns

<pauld> Pauld: outlines ideas for identifying patterns: XPath expressions asserted against a schema

<pauld> .. or possibly using a schema processor component model (SCD) or PSVI

<pauld> PaulB: wonders if assertion is good enough

pauld: we need to provide a tool that will say "yes you conform to basic patterns"

<pauld> Ajith: has to drop off

Reuse of XML Schema WG Test Suite

pauld: we have a schema test suite, containing large number of schemas

paulb: contribution were sent under a different patent policy, so reusing the schema test case might not be possible

pauld: archives are under W3C document license, which is clear

there is value is contributing test suite along with the pattern

as the goal of the group is to publish something concrete

was at the ws-addr interop workshop, very interesting to see many vendors demonstrating interop there

if would be nice to achieve the same with our patents

<pauld> ACTION: ylafon to check we're OK to reuse the XML Schema WG test suite (under the W3C document license) [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action03]

<trackbot> Created ACTION-13 - Check we\'re OK to reuse the XML Schema WG test suite (under the W3C document license) [on Yves Lafon - due 2006-01-31].

yves: support that, it proved its value in several other occasions (HTTP/WebDAV...)

jon: is it ok to require commercial tools and not prototypes?

paulb: there is benefit in showing that tools could be produced (ie: spec implementable), but it's best to release such tools

pauld: getting vendors to participate in such event would be easier if our patterns all worked with tools in productionm, the goal of the Basic patterns anyway.

pauld: like the WG to think on ideas for the test suite, identifying patterns, the assertion markup and a possible interoperability event later in the year

ADJOURNED

<pauld> thanks to Yves for scribing!

Summary of Action Items

[NEW] ACTION: items to [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action01]
[NEW] ACTION: jcalladi submit propoal based upon Erik's suggestions to soapbuilders (with Erik's permission) [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action02]
[NEW] ACTION: ylafon to check we're OK to reuse the XML Schema WG test suite (under the W3C document license) [recorded in http://www.w3.org/2006/01/24-databinding-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2006/02/08 11:12:05 $