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

RE: SOAPAction Proposal

From: Henrik Frystyk Nielsen <henrikn@microsoft.com>
Date: Mon, 7 May 2001 07:51:27 -0700
Message-ID: <79107D208BA38C45A4E45F62673A434D0297CBDC@red-msg-07.redmond.corp.microsoft.com>
To: "Martin Gudgin" <marting@develop.com>, "XML Protocol Comments" <xml-dist-app@w3.org>

>> Think of the SOAPAction value as a sort of content type and how
>> content type is used for HTML and other formats. We could have had 
>> different content type values for each HTML version - including with 
>> and without tables, scripts, math, etc. etc.
>>
>> But we didn't because the hint that this was "HTML" was good enough
>> and gave the right level of granularity for people who 
>wanted to do an
>> first order filtering/inspection of the stuff.
>
>So why don't we use text/xml+soap or some such in Content-Type
>and ditch SOAPAction?

There was a long discussion on this some time ago on this list. One
problem is that the "+" syntax maps badly onto SOAP which almost by
definition uses multiple namespaces and the other is that this would
only give half the information that SOAPAction gives. 

The half that it would give is the single bit that this is a SOAP
message but it doesn't provide any other hint about what else is in the
message so any recipient would have to parse the message to see what is
in there.

>> Making a high level hint about the content tightly bound to the
>> specific instance of a specific message is fundamentally a bad idea.
>
>OK, then let's turn the question on it's head. What *does* go
>in SOAPAction and what is it used for? ( I realise there may 
>be multiple answers... )

Specifically I would suggest that the value be a stable URI that covers
message types that are logically related like for example belonging to
the same "service".

Henrik
Received on Monday, 7 May 2001 10:52:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:01 GMT