comments from XML Schema WG

The XML Schema WG today discussed the XHTML draft now in last call
(http://www.w3.org/TR/1999/xhtml1-19990505/).  On most of the draft, we
have no comment beyond saying that we are very happy to see HTML
reformulated in XML, and think the impact of the XHTML spec may be very
high.  On two topics, however, the XML Schema WG does have comments,
which they have instructed me to convey to the HTML WG.  These are
described as follows in the minutes of the meeting
(http://lists.w3.org/Archives/Member/w3c-xml-schema-wg/1999May/0150 --
these have been submitted to the WG but not formally approved):

  We also discussed the issue of the ID and NAME attributes.  The
  issue has been raised at some length by the Linking WG, but it
  may also affect the XML Schema work, since if XHTML documents are
  served under the MIME type text/xml, it may be required that the
  XML Schema WG come up with a way to explain, at a datatype level,
  the otherwise wholly application-specific behavior of NAME
  attributes and naked fragment identifiers.  There might also be
  implications for our URI type (not agreed by all).
  
  PVB pointed out that one possible direction for the future is to
  allow multiple ID attributes, which would change the terms of the
  discussion substantially.

  The consensus of the WG was that the inconsistencies among the
  XHTML draft, the DTD, and the resolution of issues by the HTML WG
  render it very difficult to be certain that all is well.
  Since the definition of linking behavior in the XHTML may very
  likely affect the behavior of all text/xml processors, not only
  for XHTML but for all linking, we believe it's essential that the
  draft not go to PR until the Linking WG and the HTML WG have
  reached agreement on the best way to handle the NAME and ID
  attributes.  Consistency and clarity are important enough here
  that we believe the draft should not enter last call again until
  the Linking and HTML WGs have agreed, and the draft, the DTD, and
  the design documentation are all consistent with each other.
  
  There was also consensus in the WG that it would help clarify
  matters if the discussion of future plans in the XHTML document
  mentioned that the development of a schema for XHTML would be a
  plausible activity for the future, after the XML Schema work is
  further advanced.  The discussion of document profiles, in
  particular, should probably note that the specification of what
  schema modules are used in a document is a work item of the XML
  Schema WG.  We hope that the HTML WG is planning to use XML
  Schemas to accomplish this purpose; if that is not the
  expectation of the HTML WG, then the two WGs need to talk and
  align their expectations better.

Please feel free to contact me, or the Schema WG as a whole
(w3c-xml-schema-wg@w3.org) if clarification is needed, or to reply.

-C. M. Sperberg-McQueen
 Co-chair, W3C XML Schema Work Group

Received on Thursday, 27 May 1999 23:24:38 UTC