W3C home > Mailing lists > Public > www-xml-query-comments@w3.org > June 2000

Re: DTDs are parts of documents but schemas are paired with documents

From: Mary F. Fernandez <mff@research.att.com>
Date: Wed, 21 Jun 2000 14:09:17 -0400
Message-ID: <395104CD.B1B99356@research.att.com>
To: connolly@w3.org
Cc: www-xml-query-comments@w3.org
> DTDs are parts of documents but schemas are paired with documents
>
> From: Dan Connolly (connolly@w3.org)
> Date: Mon, May 22 2000
>
> *Next message: Paul Cotton: "Fwd: XSL response to Query Requirements"
>
>    * Previous message: Dan Connolly: "in-your-face URLs: please don't. (editorial)"
>    * Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
>    * Other mail archives: [this mailing list] [other W3C mailing lists]
>    * Mail actions: [ respond to this message ] [ mail a new topic ]
>
>   ------------------------------------------------------------------------

>
> Message-ID: <3929393E.337CDF98@w3.org>
> Date: Mon, 22 May 2000 08:42:22 -0500
> From: Dan Connolly <connolly@w3.org>
> To: www-xml-query-comments@w3.org
> Subject: DTDs are parts of documents but schemas are paired with documents
>
> I'm still in my first read thru, so I'm not sure how this
> impact the rest of the spec, but this looks misleading:
>
>         A well-formed document may have an associated schema, derived
>         from one or more XML Schema documents; it may have an
>         associated DTD; or it may have no schema, i.e., it is
>         ``schemaless''.
>
> http://www.w3.org/TR/2000/WD-query-datamodel-20000511/#section-Post-Schema-Validated-Infoset
>
> In XML, a DTD is part of a document. Schemas may be associated
> with documents, but they're distinct. In particular,
> it's sensible to say "I used document X with schema S1 and
> schema S2" where S1 != S2, but it's not sensible to say
> "I used document X with DTD D1 and DTD d2" because X's
> DTD is part of X. Another way of saying it:
> xml-1.0-valid is a one-place predicate over documents;
> xml-schema-valid is a two-place predicate over
> documents and schemas:
>
>         "An XML document is valid if it has an associated document type
>         declaration and if the document complies with the constraints
>         expressed in it."
>
>         -- http://www.w3.org/TR/1998/REC-xml-19980210#dt-valid
>
>         "[Definition:]  Throughout this document we use the phrase
>         schema-valid loosely to refer to a successful outcome to
>         any of the above-listed assessments of an element
>         information item with respect to a schema. "
>
>         --
> http://www.w3.org/TR/2000/WD-xmlschema-1-20000407/#validation_outcome
>
> --
> Dan Connolly, W3C http://www.w3.org/People/Connolly/
>
>   ------------------------------------------------------------------------
>
>    * Next message: Paul Cotton: "Fwd: XSL response to Query Requirements"
>    * Previous message: Dan Connolly: "in-your-face URLs: please don't. (editorial)"
>    * Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
>    * Other mail archives: [this mailing list] [other W3C mailing lists]
>    * Mail actions: [ respond to this message ] [ mail a new topic ]

Hi Dan,

Thanks for your comment.  We will modify the XML Query Data Model
document to reflect your comments.  Here is the proposed modified
text:

  We assume that an instance of the data model is derived from an
  instance of the XML Information Set after XML Schema validation.
  Such an instance is called a ``PSV Infoset'', for post
  schema-validated Infoset.

  Schema information may be associated with a document in two ways.  A
  document may refer directly to a DTD, and a document may be paired
  with an XML Schema, which may be derived from one or more XML Schema
  documents.  It is also possible for an XML document to have no DTD or
  associated schema, i.e., it is ``schemaless''.  For a document,
  schema pair, the PSV Infoset provides the schema.  For a document
  that refers directly to a DTD, we assume a mapping from the DTD to
  an equivalent schema is provided, and the PSV Infoset provides this
  equivalent schema.  For a schemaless document, a default schema is
  provided by the data model, and we specify in this document that
  default schema.  In all three cases, a schema can be provided for
  the input document and the corresponding data-model instance.

This is our proposed text, but we note that this issue is influenced
by an outstanding, last-call issue on XML Schema: LC-199.
The XML Schema: Last-Call Issues document is at:
http://www.w3.org/XML/Group/xmlschema-current/lcissues.html.

--
Mary Fernandez                    AT&T Labs - Research
Principal Technical Staff Member  180 Park Ave., Bldg 103, E243
mff@research.att.com              Florham Park, NJ 07932-0971
http://www.research.att.com/~mff  973-360-8679 FAX: 973-360-8187
Received on Wednesday, 21 June 2000 14:09:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:21:11 UTC