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

RE: text/xml for SOAP is incorrect

From: Mike Dierken <mike@DataChannel.com>
Date: Mon, 24 Sep 2001 09:57:13 -0700
Message-ID: <8E864C73E16B864BB594712EDB3C89A002C707A6@belmail2.datachannel.com>
To: xml-dist-app@w3.org


> -----Original Message-----
> From: David Orchard [mailto:orchard@pacificspirit.com]

> [...] So should the name be 
> application/getStockTicker+SAML10+crypto+dsig+soap+xml?   
> We're going to have to crack the message to get the information we need 
> anyways, so let's not warp content type just for our particular need.

You don't have to use a single name ('content-type') to encapsulate all the
values.
SOAP should have a defined mechanism for identifying within the Envelope
these values that you may want to route.

There should also be a binding specific way to identify these outside the
XML envelope - for those who believe this is the way to rapid/efficient
parsing and routing. For example, in HTTP, define new headers or use the
excellent suggestion from Graham K. (from RFC 2912)
 Content-features: (schema="http:/www.example.com/Schema-name")

Just as HTTP has headers to identify the content-encoding, etc. of the HTTP
message itself, this HTTP header can be used to identify the features
(encryption, schema/model, method, etc.) of the content specific message
within the entity body. 

This would be different for each binding, but MIME based bindings (HTTP,
SMTP) may be similar.

mike
Received on Monday, 24 September 2001 12:57:27 GMT

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