Re: MEPS Part II: SOAP

Marc Hadley wrote:

> On Apr 8, 2005, at 3:23 PM, David Hull wrote:
>
>>
>>
>>> To be honest I don't think there's much to choose between them. I 
>>> think we should just go with optional-in, optional-out and be done 
>>> with it.
>>>
>> What would [in]-[out] look like, in the HTTP binding for example, 
>> without the "in"?
>>
> It would look exactly like the existing SOAP-Response MEP.

That's what I expected.

I think this is certainly an avenue worth considering.  I'm personally 
undecided as to which approach will work best and provide the easiest 
migration path.  What /is/ clear to me now is that

    * The sender needs a way of knowing when the MEP is complete.
    * The sender needs a way to say whether it cares what comes back
    * If it does care, it needs a way of knowing whether the result was
      a message, a transport fault or nothing.

>
> Marc.
>
> ---
> Marc Hadley <marc.hadley at sun.com>
> Business Alliances, CTO Office, Sun Microsystems.
>
>