RE: Possibility of an XML Document Type

I think that there is a lot of merit to Rich's suggestion.  Some issues,
such as this one, may arise or first become prominent in the context of
XML Protocol, yet their utility is broader than just protocols.  The
ability to place one XML document within another appears to me to be
broadly useful, and consequently a matter to be addressed at the level
of XML core or XML datatypes.

I also note that the problem could be analyzed as that the binary
datatype provided by XML Schemas part II is insufficient.  It provides a
lexical space, but its "value space" is merely a string of bits, that
is, another lexical space.  The string of bits is the encoding of some
value, but the type of the value is not expressed by the facilities of
XML Schemas, neither as a facet nor via the xsi:type attribute.  An
additional facility, for example an additional attribute, is needed to
constrain the actual type of the value.

Although in earlier mail I had suggested an attribute along the lines of
xsi:mimeType="application/xml", I now believe that we should leverage
the URI namespace, and so the attribute should be more like
xsi:valueType="mime:application/xml".  I explicitly mean that any URI
identifying a type should be valid as the value of the attribute.

-----Original Message-----
From: Rich Salz [mailto:rsalz@zolera.com] 
Sent: Wednesday, October 10, 2001 9:50 AM
To: Noah_Mendelsohn@lotus.com
Cc: Christopher Ferris; xml-dist-app@w3.org
Subject: Re: Possibility of an XML Document Type

There's another possible conclusion, too: XML within XML is the job of
the XML WG, not this group.  That's an important one.
	/r$

-- 
Zolera Systems, Your Key to Online Integrity
Securing Web services: XML, SOAP, Dig-sig, Encryption
http://www.zolera.com

Received on Wednesday, 10 October 2001 13:09:27 UTC