- From: Arthur Ryman <ryman@ca.ibm.com>
- Date: Thu, 11 Nov 2004 17:44:16 -0500
- To: public-ws-desc-comments@w3.org
- Message-ID: <OFB30C6A00.60122C84-ON85256F49.007BC799-85256F49.007CE083@ca.ibm.com>
The Component Model does not enforce nesting. For example, consider two
interfaces, that each have a single operation, e.g.
interface TV {
operation PowerOn;
}
interface Radio {
operation PowerOn;
}
The TV and Radio interfaces each contain a PowerOn operation. Suppose that
the properties of the PowerOn operations are identical in both cases. The
Component Model allows there to be a single Operation component that
appears in the {operations} property of each Interface component. This
violates our intuition that operations are local to interfaces, and which
is apparent in the XML syntax for WSDL 2.0.
More seriously, this situation breaks the URI Reference specification
since that assumes we can identify any component by build up a path of
names of its ancestors. In this example, there is not unique path to the
Operation component.
I recommend that the Component Model should faithfully reflect the nesting
structure of components that is apparent from the XML syntax of WSDL 2.0
documents. This translates to the constraint that every component, except
the root Description component, have a unique parent. The parent of the
component is the component that contains the definition of the component
as in the XML syntax.
Arthur Ryman,
Rational Desktop Tools Development
phone: +1-905-413-3077, TL 969-3077
assistant: +1-905-413-2411, TL 969-2411
fax: +1-905-413-4920, TL 969-4920
mobile: +1-416-939-5063, text: 4169395063@fido.ca
intranet: http://labweb.torolab.ibm.com/DRY6/
Received on Thursday, 11 November 2004 22:44:39 UTC