See also: IRC log
<scribe> scribe: Jim
Pratul raised the concern that we do not have 50% quorum to approve minutes
MSM stated that the process document allows the Chair a fair amount of flexibility.
Pratul noted that we had quorum since Kumar had joined the call.
Chair declares that hearing no objections or concerns, the minutes for 9/13/07 are approved.
Pratul proposed that any second draft bug that is still marked as needsAgreement or needsReview at the end of this call is out of scope for the second draft.
Bug 4793 needs significant rewriting and Jim and Pratul feel it should be removed from second draft.
Decided to let the changes regarding the section on acyclic remain. The current changes related to the target* section should be rolled back.
<MSM> I propose two changes for resolving 4630
<MSM> 1) In the running text, change all references to "XML 1.0", "XML
<MSM> Schema 1.0" (etc.) to read just "XML", "XML Schema", etc., without
<MSM> specific version numbers.
<MSM> 2) For
<MSM> However, the current specification does not impose any
<MSM> restrictions on using newer versions of the XML, XML Schema,
<MSM> Schematron or XPath specifications.
<MSM> read
<MSM> However, conforming implementations MAY additionally support later
<MSM> versions of the XML, XML Schema, Schematron or XPath
<MSM> specifications.
The group agrees that the changes proposed by MSM will be made in the next draft.
The group declares consensus on the bugs listed as editorial fixes.
MSM expressed some concerns about not referring to RFC 3986 but will come back to this.
(Ref: Bug 4665)-
The group reached consensus on 4665.
The group reached consensus on 4682.
<MSM> [It's a very useful practice to say: if you DO comment on a particular issue in email, you should ALWAYS indicate the issue number in the subject line]
Valentina suggested that discussions be done in Bugzilla as this is where the editors will look when changes are made to the draft.
This relates only to discussions related to an open bug in bugzilla.
<scribe> ACTION: Valentina to write short summary of her proposal to be reviewed and approved in the next call. [recorded in http://www.w3.org/2007/09/20-sml-minutes.html#action01]
<trackbot-ng> Created ACTION-122 - Write short summary of her proposal to be reviewed and approved in the next call. [on Valentina Popescu - due 2007-09-27].
4632 Use of IRIs
<MSM> [Sandy, are you postulating a semantic difference between URIs and IRIs? At the level of abstraction we are working at, I think they are semantically the same -- IRIs are just a broader syntax for the same basic meaning]
After discussion, we are declaring consensus in accordance with the proposal Kumar put forth in email.
4635 sml:uri element name
4635 will be changed to Won't fix
4636 What should SML 3.3.1.1 say about fragment identifiers?
<Kumar> I will be back in 2 min.
<scribe> ACTION: Kumar to draft proposal on 4636 [recorded in http://www.w3.org/2007/09/20-sml-minutes.html#action02]
<trackbot-ng> Created ACTION-123 - Draft proposal on 4636 [on Kumar Pandit - due 2007-09-27].
4636 to be removed from second draft
4637 What should we do with EPR scheme?
Valentina summarized discussions she has had with Kirk:
There seems to be uncertainty about how the EPR scheme should be implemented.
Valentina suggests further investigation.
4639 Allow cycle checking on element graphs as well as documents
4639 will require further discussion.
658 How should deref() treat multiple child elements that use...
We are waiting for feedback on Sandy's proposal.
4658 will be removed from second draft.
4675 add text in section 3.2 to require that consumers and pro...
It looks like this bug incorrectly is categorized as CORE - but should be SML-IF
Pratul will send email to John asking for clarification on the intent of the bug.
4679 Revise Section 6 in SML spec
Pratul believes that this has already been done.
Sandy suggested marking this as a duplicate of 4638 Conformance Section needed.
4687 Handling of DTDs when composing an IF document
Sandy-The problem is in associating a DTD with the documents that make up the SML-IF doc, since XML requires that DTDs appear before the root element of the (SML-IF) document.
The section has been changed and the problem is still present in the spec.
<MSM> An example. The following is a well-formed (and valid) XML document:
<MSM> <!DOCTYPE foo [
<MSM> <!ENTITY bar "baz" >
<MSM> <!ELEMENT foo ANY>
<MSM> <!ATTLIST bar a CDATA 'b' >
<MSM> ]>
<MSM> <foo>&bar</foo>
<MSM> If it is embedded in an SML-IF package as described in SML-IF today,
<MSM> (a) the SML-IF package will not be well-formed
<MSM> (b) the infoset will be different (the 'foo' element in the IF package will be missing the 'a' attribute)
<MSM> Families of solutions:
<MSM> 1 require that DTD dependencies be eliminated before embedding
<scribe> ACTION: MSM to draft a rule to address handling of DTDs. [recorded in http://www.w3.org/2007/09/20-sml-minutes.html#action03]
<trackbot-ng> Sorry, couldn't find user - MSM
<scribe> ACTION: Michael to draft a rule to address handling of DTDs. [recorded in http://www.w3.org/2007/09/20-sml-minutes.html#action04]
<trackbot-ng> Created ACTION-124 - Draft a rule to address handling of DTDs. [on Michael Sperberg-McQueen - due 2007-09-27].
<MSM> 2 require that if there are dependencies on the DTD, the relevant declarations be propagated upwards
<MSM> 3 allow implementations to do either 1 or 2
4774 definitional schema documents should be preferentially used...
We will pick up with 4774 next week
The group agrees that we should request publication of the second draft.
We will make the final changes to the editorial draft and post to the web.
Kumar will notify the webmaster that we are ready to publish.