W3C

- DRAFT -

XML Processing Model WG

Meeting 176, 01 Jul 2010

Agenda

See also: IRC log

Attendees

Present
Norm, Henry, Vojtech, Paul
Regrets
Mohamed, Alex
Chair
Norm
Scribe
Norm

Contents


Accept this agenda?

-> http://www.w3.org/XML/XProc/2010/07/01-agenda

Accepted.

Accept minutes from the previous meeting?

-> http://www.w3.org/XML/XProc/2010/06/24-minutes

Accepted.

Next meeting: telcon, 8 July 2010?

Paul and Vojtech give regrets

XProc Errata

-> http://www.w3.org/XML/XProc/docs/xproc-proposed-errata

Norm: Anyone have any questions or comments about E01 and/or E02?
... Hearing none, I propose that we accept them.

Accepted.

Some discussion of what to do next; updating the errata document is the answer.

<scribe> ACTION: Norm to construct an update to the errata document pointed to from the spec and pass it off to someone who can update it. [recorded in http://www.w3.org/2010/07/01-xproc-minutes.html#action01]

Norm: Vojtech, you had a question about namespace bindings.

Vojtech: Yes, in 5.7.5, in the first list, there are rules about how to construct namespace bindings.
... The way I understand it now, if an XPath expression returns a sequence of nodes, then we use the in-scope namespace bindings off the first node if the expression returns a node set.

Norm: I think that exists so that if an expression selects a QName in content, the right namespace bindings are carried forward.

General agreement that everything is ok.

Editor's draft of XML Processor Profiles

-> http://www.w3.org/XML/XProc/docs/xml-proc-profiles

Some discussion of Paul's comment

-> http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2010Jun/0026.html

Henry: I think this would be clearer if we changed "conformant processor" to "conformant XProc processor" in that sentence.

Paul: I think that could be clearer.
... I'm happy to leave the improvements to the editor.

Norm: Paul also asks about a profile that's smaller than "minimum". I don't feel strongly about the names.
... How about "minimum", "basic", "modest", and "recommended"

Paul: That sounds good.

Norm: Anyone have concerns about these names?

None heard.

Norm: Paul's last comment is mostly editorial, but I agree.

General agreement that it should read "reading and processing" as Paul suggests.

Henry: Perhaps I should report on my action to add something about invariants
... I've started. Looking over the XML Spec again, it's not going to be as nice as I'd like.
... The best I can do for the first two profiles (which don't read any external markup) is to say things in two parts.
... For documents which are, or should be, standalone=yes and for documents which are standalone=no
... Because for documents which are standalone=no, if you don't read the external subset there isn't much you can say.
... You aren't gauranteed to get much at all.
... The most you can say is that you'll get the document element name and attributes (provided they don't contain entity references)
... But almost no one bothers with standalone="yes" and the default is standalone="no", so it'll be tricky to get right.
... Especially since processors aren't required to report unexpanded entities.
... But the other two are easier and I think we can get somewhere with them.
... The scope for variation is reduced after the external subset has been read and processed.

Norm summarizes the state of the issues list, not much progress to be made today

Any other business

Adjourned

rrsagent draft minutes

Summary of Action Items

[NEW] ACTION: Norm to construct an update to the errata document pointed to from the spec and pass it off to someone who can update it. [recorded in http://www.w3.org/2010/07/01-xproc-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/07/01 15:37:05 $