Re: Approved TAG finding: Authoritative Metadata

On Wednesday, April 12, 2006, 4:55:40 PM, Vincent wrote:

VQ> All,

VQ> The W3C Technical Architecture Group (TAG) has approved yesterday the finding
VQ> Authoritative Metadata:

VQ>     http://www.w3.org/2001/tag/doc/mime-respect-20060412

VQ> This is an update to the previously approved finding of 25 February 2004.

I see that it says:

  The TAG is working with the authors of [RFC3023] to revise section 7.1
  of that RFC, which suggests behavior regarding character encoding
  metadata that is inconsistent with this finding.

The current state of that work is now available and I would be glad of
comments. In particular, I believe that it now correctly reflects the
TAG findings in this area. Here is a summary of the changes compared to
the previous (now expired) Internet Draft:

TAG finding
-----------

I have incorporated the wording about the charset param from the TAG
finding. I used the language from the TAG finding for application/xml
charset, and for text/xml charset I strengthened the wording to say that
although optional, it MUST be used unless the encoding is us-ascii. Does
that sound right?

I also edited the summary to reflect these changes regarding charset,
and added the TAG finding to the references.

I further noted (as a consequence of the TAG wording about the charset
"where reliably known and if it agrees with the encoding declaration")
that missing out the encoding declaration should not be done. The
previous wording noted that this was sometimes done, and seemed to
slightly condone it. I think it is in practice rare to omit it, and it
should not be encouraged (unless the encoding is, in fact, UTF-8 or
UTF-16).

For application/xml, in the absence of a charset, current
implementations do what the spec now says - rely on the XML encoding
declaration, so its fine.

XPointer and its registry
-------------------------

I added a mention of the registry of xpointer schemes, and added the
registry itself to the references section.

I also changed "schemes" to "registered schemes' now that there is a
registry, and added a note that unregistered schemes SHOULD NOT be used.

Miscellaneous clarification
---------------------------

Where -01 said

   An XML document labeled as text/xml or application/xml might contain
   namespace declarations, stylesheet-linking processing instructions
   (PIs), schema information, or other declarations that might be used
   to suggest how the document is to be processed.  For example, a
   document might have the XHTML namespace and a reference to a CSS
   stylesheet.  Such a document might be handled by applications that
   would use this information to dispatch the document for appropriate
   processing.

I added "or as +xml" to be clearer.

Still to do :

SOAP and WebDAV use of text/xml
-------------------------------

Not clear what to say here; text/xml is deprecated and we don't want to
annoy them? Do SOAP and WebDAV have other Media Types that they are
transitioning towards?


Please have a look at the current work in progress:

http://www.w3.org/2006/02/son-of-3023/draft-murata-kohn-lilley-xml-02.html
http://www.w3.org/2006/02/son-of-3023/draft-murata-kohn-lilley-xml-02.xml
http://www.w3.org/2006/02/son-of-3023/draft-murata-kohn-lilley-xml-02.txt

See also the directory  http://www.w3.org/2006/02/son-of-3023/
for diffs of the xml (wrt -01.xml) and the html (wrt -01.html).

The -01 versions are in the same directory, for convenience and so I can
make diffs. The markup used is that of RFC2629; the DTD for it is in the
same directory. I used the tool at
http://xml.resource.org/
to convert the XML to the HTML and text forms.


-- 
 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 W3C Graphics Activity Lead
 Co-Chair, W3C Hypertext CG

Received on Friday, 14 April 2006 14:57:30 UTC