David,
attributes and elements aren't necessarily equivalent.
I would suggest that the schema be modified to be
consistent with the spec text. It may be important that
a foriegn attribute be added.
Cheers,
Chris
Arvola Chan wrote:
> The following is an excerpt from Section 3.2.1:
>
> The Reference element has the following attribute content in addition to
> the element content described above:
>
> � id - an XML ID for the Reference element,
>
> � xlink:type - this attribute defines the element as being an
> XLINK simple link. It has a fixed value of 'simple',
>
> � xlink:href - this REQUIRED attribute has a value that is the
> URI of the payload object referenced. It SHALL conform to the XLINK
> [XLINK] specification criteria for a simple link.
>
> � xlink:role - this attribute identifies some resource that
> describes the payload object or its purpose. If present, then it SHALL
> have a value that is a valid URI in accordance with the [XLINK]
> specification,
>
> � Any other namespace-qualified attribute MAY be present. A
> Receiving MSH MAY choose to ignore any foreign namespace attributes
> other than those defined above.
>
> The bullet in red is inconsistent with the schema (which does not have
> the #wildcard attribute). Since the Reference element already has a
> #wildcard sub-element, and we have agreed to uniformly add #wildcard
> sub-element to most key elements to provide extensibility, I would
> recommend striking out the above bullet in red.
>
> Regards,
> -Arvola
>
>
>