Re: Reqs extracted from xmlp req list

Jeff,

Some further comments on your own comments...

Jean-Jacques.

Jeffrey Schlimmer wrote:

> [...]
> > Simple applications are often characterized by message exchange
> > patterns such as one-way (or event), and two-way (or
> > synchronous) request response interactions.  The specification should
> > make such simple exchange applications as easy as possible to create
> > and to use.
> > [jeffsch: DR036: The Working Group will define a mechanism which will
> > allow a Web service to describe the following set of operations:
> > one-way messages (to and from the service described), request-response
> > and solicit-response, as described in WSDL 1.1's port types.]
>
> Correct, although I don't think DR036 covers the last sentence: "The
> specification should make such simple exchange applications as easy as
> possible to create and to use."
> [[jeffsch: I think this is adequately covered by R013.]]

Ok.

> > The WSD specification must consider the scenario where an XMLP message
> > may be routed over possibly many different transport or application
> > protocols as it moves between intermediaries on the message path.
> > [jeffsch: Use Cases Waqar is editing.]
>
> Yes, but I'd still think we need this requirement, possibly with the
> following change: s/must consider the scenario where/must support the
> scenario where/
> [[jeffsch: I don't disagree, but I'd rather not include scenarios as
> requirements given that we have a scenarios document. I expect that when
> we're happy with the scenarios, we'll go back to the requirements and
> make sure the two are in sync. Are you OK with that plan?]]

There is still some vagueness in me about the relationship between WSDL and
SOAP intermediaries, I cannot quite articulate a requirement that makes
sense right now... so I am not going to push this much further at this
point... Yes, your plan makes sense.

> [...]
> Ok, what about rewording the requirement as follows: "Must be able to
> describe accessible through one protocol and returning an answer through
> a second protocol." ?
> [[jeffsch: Is this the requirement: must be able to bind each message
> within an operation to distinct transport and wire formats?]]

Not exactly; I am thinking of the case where a different protocol is used
for the request and the response. Or maybe this is what you mean by "each
message within an operation" ?

Jean-Jacques.

Received on Wednesday, 13 March 2002 04:19:41 UTC