Attempt to clarify schema language extensibility

Hi all. 8-)

This is an attempt to clarify how WSDL can take advantage of different
schema languages.

Interaction with schema languages occurs mainly on the message part
level. In WSDL 1.1, this was the definition of a message part:

<wsdl11 section="2.3" reference="http://www.w3.org/TR/wsdl#_messages">
Each part is associated with a type from some type system using a
message-typing attribute. The set of message-typing attributes is
extensible. WSDL defines several such message-typing attributes for use
with XSD:

    * element. Refers to an XSD element using a QName.
    * type. Refers to an XSD simpleType or complexType using a QName.

Other message-typing attributes may be defined as long as they use a
namespace different from that of WSDL. Binding extensibility elements
may also use message-typing attributes.
</wsdl11>

Unfortunately, this definition is later written as 

<part name="nmtoken" element="qname"? type="qname"? />

without any reference to extensibility message-typing attributes. It
should have been written as

<part name="nmtoken" messageTypingAttribute="..." />

defining messageTypingAttribute as either unqualified 'element' or
'type' or a namespace-qualified (in other than WSDL namespace)
attribute.

In this way, one could employ RDF Schema in WSDL like in this example:

<part name="rdfPart" rdfsInWsdl:class="uri" />

where the namespace represented by the prefix rdfsInWsdl would define
what it means if a WSDL message part is of the given RDFS class.


If nobody disagrees that this was the intent of WSDL 1.1, we should file
an issue about the current draft that it doesn't catch the intent of
WSDL 1.1 and take this text as describing the real intent; the WG may
then agree either to incorporate this intent into the spec or drop this
capability.

Best regards

                   Jacek Kopecky

                   Senior Architect, Systinet Corporation
                   http://www.systinet.com/

Received on Thursday, 10 October 2002 10:14:47 UTC