Henrik Frystyk Nielsen wrote: > There is no pushing things around. We are specifically talking about > HTTP and SOAP together. The HTTP binding is silent about what happens > when using other bindings and the use of HTTP status codes is completely > orthogonal to this. Thanks Henrik. You're right, mostly. 8-) Towards the end of the discussion though, I found myself wondering if the current types of SOAP faults were sufficient. HTTP redirection would be a concept that would be difficult to hide, don't you think? What if a SOAP message is sent over HTTP, and the resource has temporarily moved? RFC 2616 sec 10.3.3 (response status 302) says that for a POST, that the user agent MUST NOT do an automatic redirect, so that suggests to me that the client should know what's going on. MBReceived on Monday, 11 June 2001 13:30:12 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:01 GMT