Re: [wsbpel] Issue - 90 - Proposal to vote

["David RR Webber" <david@drrw.info>]

Satish,

I had previously suggested a WSDL binding approach to CAM
to solve this.  If the http-binding mechanism method is not
acceptable using <from > then I suggest you instead build
that WSDL binding and use that instead.

The only issue then is error handling - since the WSDL may
return a bad status when performing the external service
call.   But I believe that is already covered off elsewhere - so
making <from > able to access a data service via WSDL
would be the other option.

Thanks, DW.

----- Original Message ----- 
From: "Satish Thatte" <satisht@microsoft.com>
To: "Dieter Koenig1" <dieterkoenig@de.ibm.com>; "Kristofer Agren"
<kagren@pakalert.com>
Cc: <wsbpel@lists.oasis-open.org>
Sent: Wednesday, April 28, 2004 11:00 AM
Subject: RE: [wsbpel] Issue - 90 - Proposal to vote


It is also unclear what the semantics of a URI used in this way would be
-- URIs don't have a native resolution mechanism so everyone is free to
interpret them as they like.  I see no way to provide semantics for
these URIs in BPEL.

This apart from the arguments Dieter gave below which I support.

Satish

-----Original Message-----
From: Dieter Koenig1 [mailto:dieterkoenig@de.ibm.com]
Sent: Tuesday, April 27, 2004 9:43 AM
To: Kristofer Agren
Cc: wsbpel@lists.oasis-open.org
Subject: RE: [wsbpel] Issue - 90 - Proposal to vote





Kristofer, the BPEL spec expresses that deployment-related issues are
out
of scope. In addition, I believe they should be kept orthogonal to the
process model.

For access to common data, the spec also has explicit statements:
"All external resources and partners are represented as WSDL services."
"BPEL4WS, builds on WSDL by assuming that all external interactions of
the
business process occur through Web Service operations."