Re: SOAP & Web arch

On 3/30/06, noah_mendelsohn@us.ibm.com <noah_mendelsohn@us.ibm.com> wrote:
> Mark Baker writes:
>
> > So while the ambiguity I pointed out is a problem with the spec
> > because it yields work which is incompatible with Web arch - as
> > demonstrated by the WS-Addressing SOAP binding's failure to populate
> > the value of the ImmediateDestination property (presumably due to the
> > WG assuming that ImmediateDestination never identifies the ultimate
> > recipient, per that ambiguity) - I agree that there's at least one Web
> > architecture-friendly use of the spec.
>
> Mark, would it them be fair to say that your concern is primarily with WSA
> as opposed to SOAP?  No doubt SOAP can be used in non-RESTful ways, but so
> can HTTP.  For example, if I choose to make up my own non-URI
> identification scheme and use it in representations sent over HTTP, the
> protocol can't stop me.  BTW: in saying this, I'm not offering an opinion
> as to whether WSA is indeed misusing Web Architecture, as I'm still trying
> to figure that out.  I'm merely pointing out that if there's a problem, it
> looks to me like it's WSA not SOAP that's the primary concern.  Do you
> agree?

Yes, I do.

And BTW, that description above about the WS-Addressing SOAP binding
not populating the SOAP/HTTP ImmediateDestination property, is
probably as succinct a description of the endPointRefs-47 issue as
I've offered.

Mark.
--
Mark Baker.  Ottawa, Ontario, CANADA.       http://www.markbaker.ca