- From: <Noah_Mendelsohn@lotus.com>
- Date: Tue, 29 May 2001 10:11:32 -0400
- To: "Asir S Vedamuthu" <asirv@webmethods.com>
- Cc: allenbr@microsoft.com, henrikn@microsoft.com, marting@develop.com, xml-dist-app@w3.org
>> a SOAP processor will >> operate on the XML infoset I agree that it should. I don't think SOAP V1.1 says this in general. Suggestion: let's make this a change or clarification in XMLP: "SOAP message formats are defined as XML Infosets. Although many bindings, such as the HTTP binding provided herein, will use the serialized form of XML provided by the XML 1.0 Recommendations as augmented by the Namespaces recommendation, other bindings can adopt other serializations. " This is a significant change but in call respects backwards compatible change to SOAP. , We have just encountered one of the many reasons why I believe this change to be important (I.e. Infoset tends to help you get the relationship between "prefixed" and "qualified" right.) I would strongly encourage those who have not to read the latest infoset draft [1], and if possible to understand that ways the recommendations such as XML schema [2] are also based on Infoset as opposed to XML itself. (In schema, both the document to be validated and the schema document used for the validation are defined as infosets, not as character streams with "<...>". Of course, for any document in "<....>" there is a corresponding infoset.) Infoset served us very well for schema. I think it is also the right way to define XMLP messages. [1] http://www.w3.org/TR/xml-infoset/ [2] http://www.w3.org/TR/xmlschema-1/ ------------------------------------------------------------------------ Noah Mendelsohn Voice: 1-617-693-4036 Lotus Development Corp. Fax: 1-617-693-8676 One Rogers Street Cambridge, MA 02142 ------------------------------------------------------------------------
Received on Tuesday, 29 May 2001 10:17:35 UTC