Conformance target of BWS40001, BWS40002, BWS40003

[Mike Edwards <mike_edwards@uk.ibm.com>]

Folks,

This is a discussion that may lead to
a new issue against the Web Service Binding spec.

Normative statements BWS40001, BWS40002,
BWS40003 all place a requirement against the SCA runtime

I don't think that this is a good construction
- and this is borne out when building testcases for these statements.

Let's examine BWS40001 as a canonical
example of all these statements:

"When the SOAP intent is required,
the SCA runtime MUST transmit and receive messages using SOAP. One or more
SOAP versions can be used. [BWS40001]"

Hmm - first, let's deal with intent.
How is the SOAP intent required?

Well, the SOAP intent must be attached
to the <binding.ws/> element (or to one of the elements in it hierarchy)
- ie. the SOAP intent is applied to the binding.ws element.
The normative statement should be clearer
about this.

Second let's deal with what transmits
and receives messages: <binding.ws/> is a subelement of either a
<reference/> or of a <service/>.
It is the reference or service that
must transmit/receive messages using SOAP. This is the thing that is measurable
- ie I can write a piece of code (non SCA) that
can receive the messages from the reference
or send messages to the service and such a piece of code can ensure that
it uses SOAP encoding.

As a result, in my opinion, BWS40001
would be better written in the following way:

"When a <binding.ws/> element
has the SOAP intent applied to it, either through direct attachment or
through attachment to another element in the hierarchy of the element,
the SCA runtime MUST ensure that the
<reference/> or <service/> which is the parent of the binding.ws
element accepts messages which use the SOAP format and transmits messages
which
use the SOAP format.  One or more
versions of SOAP can be used. [BWS40001]"

This is longer, but it is more specific
and easier to relate to the way in which the SCA runtime implementation
must behave.
It is the behaviour of references and
services which is testable in this case - writing the normative statement
in this way makes that clear.


I suggest that BWS40002, BWS40003 are
rewritten in similar ways.




Yours,  Mike.

Strategist - Emerging Technologies, SCA & SDO.
Co Chair OASIS SCA Assembly TC.
IBM Hursley Park, Mail Point 146, Winchester, SO21 2JN, Great Britain.
Phone & FAX: +44-1962-818014    Mobile: +44-7802-467431  
Email:  mike_edwards@uk.ibm.com

---

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU