Re: [ws-rx] PR Issue 13 SequenceAcknowledgement protocol responsefor AcksTo = wsa:anonymous

[Paul Fremantle <paul@wso2.com>]

Marc

In the case of AckRequested over an anon HTTP connection, I believe the 
RMD should respond directly. After all it doesn't know when the next 
HTTP connection will come in to respond on. So I think they have a point 
and we should state this.

Paul

Marc Goodner wrote:
>
> PR Issue 13, 
> http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i013 , asks if 
> an ack needs to be mapped to the original HTTP Response for the 
> message. It also suggests that maybe wsaAnonymous is not an allowed 
> value for acksTo and if so should be explicitly forbidden.
>
> I do not believe wsa:Anonymous should be barred from use in acksTo. I 
> think the spec is pretty clear that the only barred value is wsa:None.
>
> I don’t think we can address the mapping of an ack in this case to the 
> HTTP layer as the underlying binding is out of our scope. In this case 
> I think the specification already allows flexibility to send an ack on 
> the HTTP response for the original request, or on a subsequent request 
> including using AckRequested.
>
> If everyone else agrees I suggest we provide that feedback and close 
> this issue with no action.
>

-- 
Paul Fremantle
VP/Technology and Partnerships, WSO2 
OASIS WS-RX TC Co-chair

http://bloglines.com/blog/paulfremantle
paul@wso2.com
(646) 290 8050

"Oxygenating the Web Service Platform", www.wso2.com