Re: Additional syntactic restrictions

[Gavin Thomas Nicol <gtn@xxxxxxxx>]

On Saturday 08 June 2002 05:49 am, Martin Duerst wrote:
> >No, its not assuming such a design. Its conceeding that some
> >as-yet-unspecified IETF protiocol that uses XML might be used to
> >transmit an XML payload. That doesn't seem like a stretch to me.
>
> It does not seem like a stretch to me either. But it seem like
> a big stretch to me to assume that *all* as-yet-unspecified IETF
> protocols that use XML will transmit XML payloads.

The point is that it assumes embedding *arbitrary* XML content. For example, 
embedding a UTF-16 document in a UTF-8 protocol wrapper, or embedding 
something that uses CDATA sections in something that explicitly forbids them.

My point was that the protocol is the protocol, and the payload is the 
payload. You can have different restrictions on them depending on the design 
goals for each. If the protocol uses a strict subset of XML, then it can be 
processed using an arbitrary XML processor, and a subset parser will be 
sufficient to implement the specification. If the content is arbitrary XML, 
is can still be processed using an arbitrary XML processor, but may not be 
processable by the subset parser.

I see two ways of resolving that last problem, and one way requires protocol 
handlers to process XML in toto.