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

RE: SOAP and the Web architecture

From: Jones, Matthew <MJones@NetSilicon.com>
Date: Mon, 27 Aug 2001 11:52:28 -0700
Message-ID: <AD77174F26BFD411BE7B00508BFDF562108A24@newbury.netsilicon.com>
To: "'xml-dist-app@w3.org'" <xml-dist-app@w3.org>

Laird writes...
>--- Rich Salz <rsalz@zolera.com> wrote:
>> IS there a fundamental difference between sending a soap message and
>> sending form data?  If so, what is it?
>
>I think that the original objection stemmed from the fact that POST was
>conceived to augment or update the object at the URI that's being
>posted to (e.g. add to a bulletin board, put new stuff into a database,
>cause a process to start, etc.).  Form data, submitted via POST, was
>thought to be the usual way in which you'd update an object, i.e. it
>was entirely in keeping with the intended POST semantics.
>
>(For completeness: GET was to be for retrieving things, static or
>dynamic; PUT was to be for adding a new object.)
>
>I'm not sure (either) that this is always true of some SOAP
>commands--e.g. some SOAP requests are decidedly more
>give-me-something-oriented than update-this-resource-please-oriented.
>

It never seemed to me that post was particularly for updating an object.
Perhaps that is implied by a content type of multipart file upload, but for
url form encoded I wouldn't infer update-this-resource-please-oriented.
Many search requests and other requests (ie stock price) use post.  I always
inferred that post was supposed to be more general than get.  I have read
that get requests should be limited to 256 bytes as that was all that some
servers accept (of course that isn't true in practice at least now) and that
because of such perceived limitations of get that post was preferred.

Matthew Jones
mjones@netsilicon.com
Received on Monday, 27 August 2001 14:53:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:03 GMT