W3C home > Mailing lists > Public > public-xml-core-wg@w3.org > September 2004

Minutes for XML Core WG telcon of 2004 September 8

From: Paul Grosso <pgrosso@arbortext.com>
Date: Wed, 8 Sep 2004 12:01:39 -0400
Message-ID: <F13E1BF26B19BA40AF3C0DE7D4DA0C034FD3C9@ati-mail01.arbortext.local>
To: "XML Core WG" <public-xml-core-wg@w3.org>

 Anjana  xx:22
 John  xx:11

[10 organizations (11 with proxies) present out of 12]

Lew, proxy to the chair

Absent organizations
W3C, with regrets
Lew Shannon, with regrets, proxy to the chair

> 1. Accepting the minutes from the last telcon [3] and
>    the current task status [2] (have any questions, comments,
>    or corrections ready by the beginning of the call).


> 2. Miscellaneous administrivia.
> 3. Problem with xml:space in the Schema document for the XML namespace
> Masayasu Ishikawa <mimasa@w3.org> sent us email on this at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Jul/0019
> Henry found that Mimasa was not correct about not being able
> to  make xml:space fixed.  Henry replied at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Aug/0013
> But it remains open whether we should change the default (or
> change the fact that it is defaulted).
> CONSENSUS to remove the default for xml:space from the schema
> for the xml namespace.
> ACTION to Henry:  Make the actual change once we've figured
> out how/when to announce this.
> ACTION to JohnC:  Make an announcement to xml-dev that we plan
> to make this change and request any feedback be sent to xml-editor.

Attempted by John.  Henry got one through to xml-dev, so Done.

> ACTION to Henry:  Ditto to xml-schema-dev and to chairs.


> 4. XML errata.  The published 1.0 errata document is [8], the
>    published 1.1 errata document is [9], and the NEW PUBLIC
>    Potential Errata (PE) document is [7]. 
> 5. Namespaces in XML.
>   ACTION to Richard:  Produce a draft for NS1.0 2nd Ed.
> 6. Xinclude CR was published April 13 at:
>    http://www.w3.org/TR/2004/CR-xinclude-20040413
>    The updated test suite cover page is at
>    http://www.w3.org/XML/Test/XInclude/ 
> The PR-ready draft is at:
> http://www.w3.org/XML/Group/2004/07/PR-xinclude/
> The public DoC (aka latest issues list) is at:
> http://www.w3.org/XML/2004/07/ExIT-xinclude/issues.html
> [Note: The Director view displays incorrectly in IE6.0.]
> At 
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Jul/0025 
> Richard had sent a format for submitting test reports and an XSLT 
> to convert the report to an HTML page. 
> ACTION to Richard:  Add a test for xml:lang to the test suite.

Done--it's eduni-3.

DV did run but did not pass this test--he doesn't support xml:lang.

ERH does suppport xml:lang, but he hasn't run this test.  He ran
his own tests for xml:lang.

> Richard put up results for ERH and himself:
> http://www.w3.org/XML/Group/2004/08/xinclude-implementation/re
> port.html
> Paul and Richard got some more info from Elliotte including his
> test suite.  It turns out that we'll need some written permission from
> Elliotte to include his test suite--Paul will look into that.
> Meanwhile:
> ACTION to Richard:  Check out Elliotte's results and sanity check that
> he does support xml:lang and accept-*; assuming he does, add some kind
> of entry/note to our implementation report document saying 
> that he does
> so that we can "document" that we have an implementation of these
> features.

Elliotte sent us 94 tests in his suite; Richard disagreed on 19 of them.
Some disagreements are due to things Richard doesn't support, but others
might be real disagreements.

ACTION to Paul:  See about getting approval to incorporate ERH's tests
into the W3C test suite.

ACTION to Richard:  Go through ERH's results and figure out if/where
there are any real issues.

ERH's tests raised an issue about in-scope namespaces for Richard
The included content (after inclusion) might have extra namespaces 
in scope from ancestors of the xinclude element in the including 
document (unless they have been turned off using NS 1.1).  

Is this to be allowed by/authorized by XInclude or not?

Jonathan: The "one true infoset" doesn't have these namespaces in 
scope, but whether a given serialization of such an infoset reflects
this completely accurately is not necessarily an XInclude issue.

We seem to agree that it's alright for the included nodes to have
in-scope namespace nodes that they did not have in their original

CONSENSUS:  We should add something like "implementation MAY 
include extra namespace nodes...."

ACTION to Richard:  Suggest some actual wording to go into the spec.
Send the wording to the list.

> There is email from Dan Connolly re:
> how does XInclude mix with XML Schema? XSLT?
> at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0014

Why does XSLT have its own inclusion mechanism (timing, history), 
and can one use xinclude with XSLT and XML Schemas?

Both XSLT and XML Schema includes have added semantics over 
simple XInclude (and external entity refs).

But there is no reason one couldn't use XInclude instead of
external entity refs in XSLT and XML Schema.  In this sense,
an XSLT or XSD document is just an XML document, so XInclude
should work as in any XML document.

ACTION to Norm:  Respond to Dan.

> Daniel sent email about his results at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0015
> but I'm not sure how to present this in our implementation 
> feedback, so:
> ACTION to DV:  Provide a table giving results (using 
> Richard's files) of
> running the test suite on your implementation.

DV sent a hand-updated HTML at

> Paul sent an updated draft PR request at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0013
> Remaining ACTIONs are:
> a.  Richard:  Copy the implementation report into public 
>               space and email the new URL to the WG.

ACTION continued; put it at:

> b.  Daniel:  Provide us with an implementation report asap.


> c.  Paul:  Get from Elliotte a definitive statement about
>            his support of xml:lang and accept/accept-language.
>   This morphed into the above action to Richard to augment the
>   implementation report with such a statement.

ACTION to Richard:  Double check to see that Elliotte does have
an implementation of xml:lang and accept/accept-language.

> d.  WG:  Approve this WD to go to PR.

The WG has CONSENSUS to send
(with minor editorial tweeks) to PR.

> e.  Paul and Norm:  Create the final version in WG space.
> f.  Paul:  Generate and send the actual PR request.
> g.  Henry:  Help arrange a PR call.
> Paul ran pubrules and things look good, but since last minute
> changes will require another pass:
> ACTION to Paul:  Do a final pass on the status section, 
> pubrules, etc.

ACTION to Paul (perhaps with help from Norm and Henry):  Create
the final version of the spec in WG space, do pubrules checks,
and generate and send the PR request.

> 7. xml:id.
> Relaxing the constraint that there be one ID per element.
> ---------------------------------------------------------
> http://lists.w3.org/Archives/Public/public-xml-id/2004Apr/0012
> We want to make sure that the xml:id spec is agnostic wrt whether 
> there is more than one thing of type id on one elemnt, as this is 
> a property of the validation mechanism.  
> Currently, the spec says nothing in this regard, so it is, in fact,
> agnostic.  Therefore, we have no action.
> So the reply to the comment is that the xml:id spec has no such
> constraint, so there is nothing to relax.
> CONSENSUS to add a note to the spec on this.
> ACTION to Norm:  Add the note and reply to the commenter.
> Norm has collected the xml:id issues, now (public) at:
> http://www.w3.org/XML/2004/xml-id/issues.xml
> and put a new version of the draft at
> http://www.w3.org/XML/Group/2004/xmlcore/xmlid/xml-id.html
> ID Strictness 
> -------------
> Commentor figures that non-validating parsers shouldn't have 
> to check for xml:id validity.
> Currently, conformance to xml:id does require non-validating 
> parsers to check for xml:id validity; of course, no parser is 
> required to conform to xml:id.
> Richard send email outlining our options at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Aug/0024
> He outlines options of strict, moderate, lax id checking.
> Richard would like to know that ids, for example, don't 
> contain spaces.
> Then we started asking if id values should be NCName or Name.
> Tentative CONSENSUS that we would do "moderate" where id values 
> should be NCName (with some uncertainty).
> We left open what kind of error it would be if the above isn't 
> the case.
> Norm sent email with his latest proposal:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0003
> and there was some follow up email discussion.

We tried to have this discussion again in the last few minutes,
but did not reach finality on the language.  To be continued next week.

> 8.  XML Profile.  The TAG (via Norm) asks about progress on this:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0004
> [1] http://www.w3.org/XML/Group/Core
> [2] http://www.w3.org/XML/Group/Core#tasks
> [3] 
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0002
> [7]
> http://www.w3.org/XML/2004/02/proposed-xml10-3e-and-xml11-errata.html
> [8] http://www.w3.org/XML/xml-V10-3e-errata
> [9] http://www.w3.org/XML/xml-V11-1e-errata
Received on Wednesday, 8 September 2004 16:01:55 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:16:33 UTC