ISO/IEC JTC 1/SC 34N0566
ISO/IEC JTC 1/SC 34
Information Technology --
Document Description and Processing Languages
|TITLE:||Minutes of SC34/WG1 Meeting, 13/14th November 2004|
|SOURCE:||Mr. Martin Bryan|
|STATUS:||Minutes of working group meeting|
|DISTRIBUTION:||SC34 and Liaisons|
|REFER TO:||N0555 - 2004-11-07 - Agenda for WG1 meeting, Washington, November 13th/14th 2004|
Dr. James David Mason
(ISO/IEC JTC 1/SC 34 Secretariat - Standards Council of Canada)
Crane Softwrights Ltd.
Kars, ON K0A-2E0 CANADA
Telephone: +1 613 489-0999
Facsimile: +1 613 489-0995
Network: [email protected]
Minutes of SC34/WG1 Meeting, 13/14th November 2004
Diederik Gerth van Wijk
The meeting convened at 9am EST on 13th November and ended at 4:20pm on 14th November.
The proposal made to add the svrl: namespace to Annex D of Part 3 was discussed. It was proposed that the DTD needed to be changed to RNC if namespaces were to be used validly in the examples as the proposed using of xmlns as an attribute in the DTD would conflict with the proposal for handling namespaces in DTDs proposed for Part 9 in Cambridge.
The proposal from Rick Jelliffe that Part 9 should be based on Part 2 was discussed and rejected by the committee, who felt that at a minimum namespace support and datatype validation requirements should be identifiable using PI's associated with the DTD. It is noted that PIs form part of the XML Infoset.
The committee undertook a detailed review of Part 1 prior to its submission for national body review. A revised version was created.
The committee reviewed submissions on Part 7 submitted by email in the form of uncommented RNC files from Japan during the course of the meeting and made a number of comments on its omissions, requesting that a formal description of the proposal be prepared so that it can be more accurately understood.
Comments on Part 8 were resolved as followed:
1) It should be possible to use any XPath statement that returns a node set to identify the node to be renamed.
2) The info set passed to the validating schema should be that required for validation (the newly mapped name). A PI associated with the start tag should identify the XPath statement that identifies the node being validated in the document instance.
3) It should be possible to map names to default entities, but not to reassign the default names to another default name.
4) PI mapping (and other mapping) can be done from outside (e.g. maps declared in Part 10)
5) Property renaming possible, but need to state only where property-name=value convention is being used.
6) While the remove function may be useful it should not be in a part devoted to renaming. If DSDL is allowed in Part 10 then it is not needed here.
7) XInclude requires fragments to be well-formed so does not provide a generalized mechanism. (not "set of elements" – only a "an element")
It was agreed that we would explore the possibility of using existing pipelining languages to implement Part 10, making it a TR on the use of one or more of the existing languages rather than the specification of a new language.