- From: Mark Baker <distobj@acm.org>
- Date: Wed, 3 Apr 2002 15:42:35 -0500
- To: Henrik Frystyk Nielsen <henrikn@microsoft.com>
- Cc: Mark Baker <distobj@acm.org>, noah_mendelsohn@us.ibm.com, "Williams, Stuart" <skw@hplb.hpl.hp.com>, xml-dist-app@w3.org
Henrik, On Tue, Apr 02, 2002 at 09:37:26PM -0800, Henrik Frystyk Nielsen wrote: > > >Now I'm confused. 8-/ I used to believe this, when I thought that > >faultHint was authoritative, but now I wonder how you could say that > >when, AFAICT, nowhere in the binding does it distinguish between a > >fault received on a 200 response, and one received on a 500 response. > >Both end up in the success state, and only the faultHint distinguishes > >one from the other. > > I thought we said that the former is simply broken - it won't happen if > the implementation is conformant with the SOAP HTTP binding? Yes, that would be better than the status quo. It still doesn't address my issue completely, as I would expect that SOAP implementations would be "liberal in what they consume". So, without a spec that says anything to the contrary, they would likely treat a fault on 200 as a fault. And that would yield problems for HTTP intermediaries, which my company produces. Perhaps we should just call this a non-chameleon binding (as David just mentioned), and agree that it isn't suitable for use with all HTTP intermediaries. At least that way we won't have people thinking that it was. I would agree to close issue 192 if we said that. MB -- Mark Baker, Chief Science Officer, Planetfred, Inc. Ottawa, Ontario, CANADA. mbaker@planetfred.com http://www.markbaker.ca http://www.planetfred.com
Received on Wednesday, 3 April 2002 15:50:23 UTC