XML Schema Patterns for Databinding Working Group Teleconference

25 Mar 2008


See also: IRC log


Jon Calladine (BT)
George Cowe (Origo Services Limited)
Paul Downey (BT)
Yves Lafon (W3C)



minutes from 18th approved


gcowe: we shouldn't have directories with "edcopy" in the URI

pauld: we'll take a snapshot of the collection and testsuite report and publish when we go to the Director for PR
... believe I have completed the editing for Last Call

yves: we need to highlight moving directly to PR in the status section

pauld: done!


jonc: so what about "Block Default"? http://lists.w3.org/Archives/Public/public-xsd-databinding/2008Mar/0015.html

jonc: it's an innocuous default pattern

pauld: we have a statement in the status section about not introducing new elements and attributes
... have you tested them? "./@blockDefault and not(./@blockDefault)" looks strange

gcowe: happy with the first option, seems an unusual thing to do

pauld: there is precedence, e.g nillable

jonc: and we have others globalComplexTypeBlock and globalElementBlock

pauld: this seems better ./xs:element[@block=""]/(@block)

gcowe: blockDefault and finalDefault have been skipped on all of the tools

pauld: can we move these to advanced?

gcowe: we don't have much evidence

jonc: doesn't seem dangerous, but without testing ..

gcowe: we don't use it, and my guess is if we test this it'll be advanced

RESOLUTION: move GlobalComplexTypeBlock, GlobalElementFinal, GlobalElementBlock to advanced


pauld: George points out that NonIdentifierPattern is still basic
... document here: http://localhost/2002/ws/databinding/edcopy/basic/basic.html

gcowe: we're hitting this due to names longer than 32

pauld: we discussed this under ISSUE-10

gcowe: we don't have very good examples in this area

pauld: we should add some more tests
... we could split the pattern to be "symbolic name" and "long symbolic name"

<gcowe> ISSUE 10 discussed in this thread

<gcowe> http://lists.w3.org/Archives/Public/public-xsd-databinding/2006Sep/0002.html

gcowe: 31 is the magic number

pauld: worried, because we're sure non-ASCII names is an issue, but the length?
... let's create some examples

gcowe: worried by the number of examples in this area

pauld: we could create a little test suite for this alone

gcowe: we've tried long names in C# and Java without issues, but obviously C might ..

pauld: OK we'll work on this

Detection Service

gcowe: thinking about adding highlighting, also been thinking about rerunning test suite

Last Call

pauld: we're mostly done

yves: links to Test Suite and Collection?

pauld: will do so


pauld: pickup in two weeks

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/07/01 15:18:14 $