W3C home > Mailing lists > Public > public-xml-core-wg@w3.org > May 2005

Minutes for XML Core WG telcon of 2005 May 4

From: Paul Grosso <pgrosso@arbortext.com>
Date: Wed, 4 May 2005 12:05:52 -0400
Message-ID: <F13E1BF26B19BA40AF3C0DE7D4DA0C03045B7411@ati-mail01.arbortext.local>
To: <public-xml-core-wg@w3.org>
Cc: "Rich Salz" <rsalz@datapower.com>

Attendees
---------
 Paul 
 Glenn xx:09
 Arnaud
 Dmitry  xx:05
 Norm 
 Leonid
 Richard
 Henry
 François 
 Daniel
 John xx:21

Rich Salz, Datapower, invited to this telcon
	for agenda item 10.

[9 organizations (9 with proxies) present out of 10]

Regrets
------- 

Absent organizations
--------------------
Lew Shannon


Norm sends regrets for next week.

> *************************************************************
> *                                                           *
> * We will take agenda items 10 (with Rich Salz attending)   *
> * and 11 first (after administrivia) in this week's telcon. *
> *                                                           *
> *************************************************************
> 
> Agenda
> ======
> 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).

Accepted.

> 
> 2. Miscellaneous administrivia and document reviews.
> 
> The new XML Core WG charter has been approved.
> The Call for Participation is out, and everyone on the WG
> has to have their AC rep submit their name as a member in
> the rechartered WG by May 20th:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0006
> 
> Richard reviewed the 
> XPath 2.0/XQuery 1.0 Data Model document that is at:  
> http://www.w3.org/TR/2005/WD-xpath-datamodel-20050211/
> Richard's review is at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0014
> 
> There is also an issue about what the types are in the data model:
> the schema types or another system that is similar.  Henry and
> Richard point out the type hierarchy in this data model spec
> is not quite the same as in the XML Schema spec.
> 
> ACTION to Richard:  Augment the earlier email with respect
> to the above issue and send them in as XML Core WG comments.

Done.


We've received a response from the QA group on our
comment about QA Framework.  See Paul's message
summarizing this at
http://lists.w3.org/Archives/Public/public-xml-core-wg/2005May/0004

NOT discussed this telcon.

> 
> 3.  XLink update.
> 
> The first WD of XLink 1.1 has been published:
> http://www.w3.org/TR/2005/WD-xlink11-20050428/
> 
> The Issues/DoC list is at:
> http://www.w3.org/XML/2005/04/xlink11/wd-status/
> 
> 
> 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]. 
> 
> We had a question about the XML Test Suite arise; see
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Mar/0037
> 
> Awaiting response from Richard.
> 
> 
> 5. Namespaces in XML.
> 
> Richard suggested we take NS 1.1 and revert the two 
> substantive changes (IRI and undeclared namespaces) 
> to create NS 1.0 2nd Ed. The WG has consensus to do 
> that, and we got approval from the team to do so.
> 
> Ongoing ACTION to Richard:  Produce a draft for NS1.0 2nd Ed.
> 
> We note that the IRI spec is now finished-RFC 3987-so 
> we have to issue an erratum for NS 1.1 for this.  We
> discussed some details of this under the XLink discussion:
> http://www.w3.org/XML/2005/02/xml-f2f-20050303-minutes.htm#xlink
> Briefly, 3987 does have some wording (the "MAY" paragraph) 
> about what used to be called unwise characters.  For the 
> NS 1.1 erratum, the MAY paragraph doesn't apply since 
> namespace names cannot have the unwise characters.  (The 
> MAY paragraph will be needed for XML 1.* system identifiers.)
> 
> ACTION to Richard:  Process an erratum to NS 1.1 to
> refer to RFC 3987: http://www.ietf.org/rfc/rfc3987.txt
> 
> 
> 6. Xinclude Rec was published 2004 December 30 at:
>    http://www.w3.org/TR/2004/REC-xinclude-20041220/
> 
> Our XInclude potential errata document is at:
> http://www.w3.org/XML/2005/01/proposed-xinclude-errata
> 

Note: Henry just corrected the XInclude namespace doct.

> See
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Mar/0029
> for our PE document which is awaiting updating by DV.
> 
> ACTION to DV:  Update the XInclude PE document with the resolutions.
> 
> 
> 7. xml:id.
> 
> The CR was published (2005 Feb 8) at
> http://www.w3.org/TR/2005/CR-xml-id-20050208/
> 
> The (public) xml:id LC issues is at:
> http://www.w3.org/XML/2004/xml-id/lc-status/status-report.html
> The LC DoC is at:
> http://www.w3.org/XML/2005/01/xml-id-lc-doc.html
> Our implementation report is at
> http://www.w3.org/XML/2005/01/xml-id-implementation.html
> We have a test suite cover page at
> http://www.w3.org/XML/Test/xml-id/
> 
> Norm sent some email at
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Mar/0023
> and a sample of his implementation feedback at
> http://www.w3.org/XML/2005/01/xml-id/xmlidfilter-report
> 
> Richard put his implementation report at
> http://www.w3.org/XML/2005/01/xml-id/rxp-report.html
> 
> Richard had some questions on Norm's latest test suite.
> 
> On the last test, Norm fails because XSLT can't do it.
> Norm gets a space in it that shouldn't be there.  When
> Richard runs it, he gets the empty string for the result.
> 
> ACTION to Norm:  Investigate what should happen on this
> last test.

Norm:  XSLT's id function can't test the issue about
whitespace in an id.

Richard:  But Norm and Richard got different results
out of their XSLT processors.  We thought this might
have to do with being normalized twice (or something).

Norm:  OK, that's what I need to investigate.

ACTION to Norm:  Investigate why Richard and Norm are
getting different results from their XSLT processor
for the last test.

> 
> ACTION to DV:  Run your implementation on the 
> test suite and produce some feedback report.

DV did; results at:
http://veillard.com/xmlidresult.html
which Norm will pick up and put into W3C space.

ACTION to Norm:  Get DV's results into W3C space.


> We discussed changing wording about errors so that an xml-id
> processor doesn't need to report errors *to the application*.
> 
> In Section 6 Errors, we currently say:
> 
>   A violation of the constraints in this specification
>   results in an xml:id error. Such errors are not fatal,
>   but must be reported by the xml:id processor to the
>   application invoking it.
> 
> ACTION to Richard:  Suggest some rewording for this and
> pass it by ERH.
> 

ACTION continued.

> Paul sent email to the CSS WG about xml:id:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0091
> and there have been a couple responses.
> 

The CSS group was reasonably positive about the idea
and is planning to add wording that covers xml:id:
http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0092

> 
> 8.  Associating stylesheets--awaiting TAG action.
> 
> Henry noticed that the HTML CG has run into the same issue.
> There is an interaction between media types and secondary
> resource, and there appears to be no consensus on the HTML CG
> as to what should be the case.
> 
> Henry asked the HTML CG if they felt this issue should be
> taken to the TAG, but Henry isn't getting a single voice 
> out of the HTML CG.  He will continue to work on this.
> 
> ACTION to Henry:  Continue to see if this issue should
> be brought to the TAG.
> 
> 
> 9.  absolutivity of [base URI]
>     Norm has asked a question about the absolutivity of [base URI]:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Feb/0031
> 
> We discussed this at our f2f:
> http://www.w3.org/XML/2005/02/xml-f2f-20050303-minutes.htm#base-uri
> 
> We have CONSENSUS that base URIs are always absolute. 
> 
> Then we had a further issue about base URIs in the infoset.
> 
> Richard sent email to www-tag on this:
> http://lists.w3.org/Archives/Public/www-tag/2005Apr/0077
> 
> ACTION to Richard:  Review the responses.
> 
> 
> 10.  XML Validity and DTD dependence.
>   Rich Saltz started the discussion at
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0026
> and there have been several sub-threads.  
> 
> Paul has invited Rich to this week's telcon.

Rich explained that validity is important to him.
He mentioned id validity, but he says there are
other validity constraints in the XML spec that
are important to him.  But since DTDs are not
namespace-aware, and since he is interested in
validity in the presence of namespaces, he is
looking for a new definition of validity that works
without DTDs.

Richard:  The XML spec defines DTD-validity, but just
calls it "validity".  It appears that Rich is interested
in something other than DTD-validity, but then we shouldn't
call it (unqualified) "validity".  

Rich wants some kind of "beyond just DTD validity" that
allows him to know if his application should be able to
handle the current document or can just say "bad document".

Richard:  Different kinds of schemas can impose different
contraints than others, so defining generic validity could
be difficult.

Rich:  Start with the validity constraints from XML 1.0
that aren't DTD related, such as "element valid" in terms
of that fact that its children match the content model.

Norm:  You mentioned uniqueness of ids.  Are there other
constaints in which you are interested?  Would just getting
uniqueness of ids satisfy your biggest use case, because in
that case you will get that with the xml:id spec.

Rich:  Yes, that is the biggest one.

Richard:  Are you also concerned about idref completeness,
because xml:id doesn't address that?

Rich:  Yes, but secondarily.

Richard:  The W3C came up with XML Schemas to address the
fact that DTDs don't handle namespaces, so in what way
doesn't XML Schemas address your requirements?

Rich:  I want to be able to associate validity with a document.

Richard:  But you need to say valid with respect to some schema,
you can have validity in the absence of any schema.

Rich:  That's the thing I'm questioning.  Isn't there some way
to define validity in the general sense without any specific
schema?

DV:  Validity is a contract based against a schema.  Well-formedness
is a property of the document, but validity is a contract against
a set of constraints that are defined outside the document.

Richard:  How do you know what things are ids?

Rich:  I have a partial list of things I'm going to treat as id 
attributes.

Henry:  You can write an XML Schema that just validates uniqueness
of ids for a given list of id attributes provided they are prefixed.

Rich gives an example of a SOAP application that has built-in
knowledge about the schemas of documents it is using.

Richard:  If you want things with only partial knowledge to be
able to validate ids, xml:id should give you what you need.

John:  XML 1.0 validity is a property of the "whole document"
which includes the DTD.

Richard:  I think we understand each other better now, but I'm
still not sure what we can do.

Rich:  I think it's up to me now to propose something.

Richard:  We must leave the lowest level--XML--untouched.
If there is a new definition of some kind of validity, it
would have to be at a higher level.

Richard:  If you are only interested in ids in a specific
namespace (e.g., ids in the SOAP wrapper), then you can 
write an XML Schema that validates them.  The XPointer
Framework spec talks about ways things can be determined
to be ids.

Any followup is now in Rich's court.

> 
> 11.  XInclude, schema validity-assessment, xml:base and xml:lang
> 
> Henry kicked this off at:
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2005Apr/0039
> We didn't discuss this topic much ourselves, but Norm and Henry 
> both lean toward making it an issue for the XML Schema WG, so we 
> are waiting to hear from them.
> 
> We will take this near the beginning of the call next week
> so that we have more time to discuss it.  People should read
> the above URL before the telcon.

XInclude requires xml:base fixup with adds xml:base
attributes to a document.  This causes problems 
validating the result against the original schema
if that schema doesn't mention xml:base.

Norm wants the XML Schema group to have a mode that
says "just assume all xml:* attributes are okay".

Henry points out we even have problems with validation
against DTDs in this case.

It was suggested that we add to the XInclude spec:
"An XInclude processor may, at user option, suppress
xml:base and/or xml:lang fixup."

We will come back next week to see if we are willing
to add that phrase.

> 
> paul
> 
> [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/2005Apr/0090
> [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, 4 May 2005 16:07:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:32 GMT