W3C home > Mailing lists > Public > xml-names-editor@w3.org > December 2002

RE: FW: XML Query WG Feedback on Sept WD of Namespaces in XML 1.1

From: Michael Rys <mrys@microsoft.com>
Date: Thu, 5 Dec 2002 11:59:37 -0800
Message-ID: <5C39F806F9939046B4B1AFE652500A3A039BE9DC@RED-MSG-10.redmond.corp.microsoft.com>
To: "Paul Grosso" <pgrosso@arbortext.com>, "Kay, Michael" <Michael.Kay@softwareag.com>
Cc: "XML Core WG" <w3c-xml-core-wg@w3.org>, <w3c-xml-query-wg@w3.org>, <xml-names-editor@w3.org>

I think that we can avoid revving the Infoset along with XML 1.1 only
if:

1. XML 1.1 describes a true superset of XML 1.0
2. XML 1.1's superset is not adding new concepts but only adds to the
value space of the information items (ie, undefining namespaces,
allowing more character information items) or is purely syntactical
(U+0002 has to be entitized).


In any other case, the Infoset needs to be rev'ed as well. I don't think
it is acceptable to have Infosets that combine 1.0 and 1.1 information
items if the requirements above do not hold. 

Best regards
Michael

> -----Original Message-----
> From: Paul Grosso [mailto:pgrosso@arbortext.com]
> Sent: Thursday, December 05, 2002 7:25 AM
> To: Kay, Michael
> Cc: XML Core WG; w3c-xml-query-wg@w3.org; xml-names-editor@w3.org
> Subject: RE: FW: XML Query WG Feedback on Sept WD of Namespaces in XML
1.1
> 
> 
> [I am replying to Michael K's reply to Richard's reply to Michael R's
> reply to Richard's reply to the Query WG Namespace 1.1 comments.
> Richard's reply had the wrong address for the XML Core WG, so MK's
> reply did too.  paul]
> 
> >From: "Kay, Michael" <Michael.Kay@softwareag.com>
> >To: Richard Tobin <richard@cogsci.ed.ac.uk>, Michael Rys
> <mrys@microsoft.com>
> >Cc: w3c-xml-query-wg@w3.org, w3c-core-wg@w3.org,
xml-names-editor@w3.org
> >Date: Thu, 5 Dec 2002 13:02:48 +0100
> >Subject: RE: FW: XML Query WG Feedback on Sept WD of Namespaces in
XML
> 1.1
> >
> >>
> >> Infosets produced by parsing XML 1.1 documents will have the
> >> [version] property of the document information item set to
> >> 1.1 (we don't even need an amendment to the Infoset spec for that).
> >>
> >> > (Note that this is not relevant to Namespaces 1.1 but XML 1.1).
> >>
> >> I think it should really be considered a comment on the
> >> Infoset revision.
> >>
> >
> >I guess I'm not going to persuade anyone, but I think it is really
> >unfortunate that the Core WG has decided to update the XML and
Namespaces
> >specs without making them reference the Infoset normatively. The
result
> is a
> >missed opportunity to clear up the confusion as to what parts of an
XML
> >document are information-bearing and what parts are not.
> 
> This is really more a comment on XML 1.1 and only tangentially on
> Namespaces 1.1.
> 
> The issue is one of requirements scope.  We made it clear from the
start
> that we were limiting the scope of XML 1.1 and Namespaces 1.1 very
> severely.
> The kind of change you suggest would take much more work and would
likely
> have to be an XML 2.0 which would in turn open it up to many more
desired
> changes which would in turn make it take exponentially longer to the
point
> where the window of opportunity could be completely missed.  The XML
Core
> WG felt that it was preferable to bite off a smaller task and get it
done
> sooner.
> 
> 
> >At some stage we need to invert this whole edifice: the InfoSet data
> model
> >should be the primary specification, and the XML and Namespaces specs
> >(hopefully merged) should merely describe one possible interface for
> >creating an InfoSet.
> 
> Many people agree with you here.  In fact, when the XML Core WG last
> discussed possible futures at our f2f last February, such
reorganization
> of the specs was high on the list of possible things to do.
> 
> The issue is always "at which stage".  It would have been best if done
> five years ago when XML was developed.  At this point, though, one has
> to weigh the benefits of cleaning up specs and reducing confusion--but
> not really adding any benefit for end users--against the huge cost of
> the actual effort.
> 
> paul
> 
> 
Received on Thursday, 5 December 2002 15:00:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:43 GMT