W3C home > Mailing lists > Public > xml-dist-app@w3.org > May 2001

RE: SOAPAction Proposal

From: Andy Neilson <ANeilson@webplan.com>
Date: Thu, 10 May 2001 11:44:13 -0400
Message-ID: <39CF16D2A30FD5119E83000629F61CA90A94D5@wpkanex.webplan.com>
To: xml-dist-app@w3.org
My problem with SOAPAction (as it is defined in SOAP 1.1) is the use of the
word MUST, as in:

    An HTTP client must use this header field when issuing a SOAP HTTP

I can't see why this is required. While I can imagine that SOAPAction is
useful in some situations, I can't see any basis for claiming that it is
always necessary. If a provider of a service requires the SOAPAction header,
then they can describe their service as requiring it (e.g., in WSDL). For
those of us that have no use for SOAPAction, I see no reason why my HTTP
SOAP clients should be compelled to include the SOAPAction header.

We don't use SOAPAction in any of our SOAP implementations.

The only basis I can see for requiring the SOAPAction header is some attempt
at forward compatibility (i.e., in case you need to use it someday). If I
did need SOAPAction someday, then I would simply publish a new endpoint that
described the requirement for the SOAPAction header.

I suggest that the SOAP 1.1 requirement that the SOAPAction header "MUST" be
toned down to a "MAY".

Andy Neilson
Received on Thursday, 10 May 2001 11:48:09 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 22:01:13 UTC