- From: Frystyk <frystyk@w3.org>
- Date: Fri, 20 Apr 2001 15:39:27 -0700
- To: <dick@8760.com>, <soapbuilders@yahoogroups.com>
- Cc: <xml-dist-app@w3.org>
>All the web servers I've used (IIS, Apache, Netscape) provide >access to the HTTP headers and the SOAPAction contents are >easily accessible to a CGI, ISAPI, NSAPI or servlet program. >Once a message broker has the SOAPAction data it can use the >information without any further expansion (it's relative to >the Host/HTTP POST URI of the message broker). This actually makes no difference - the problem is that ebXML can say nothing about these relative URIs: neither what they identify nor how they relate. It would be the same if HTTP tried to say that all URIs that have the word "cgi" somewhere in the string is a cgi script. The model behind URIs does not allow this. In order for the ebXML spec to say anything about the intent of the message, it has to use an absolute URI. What's the problem in using an absolute URI? Henrik
Received on Friday, 20 April 2001 18:40:08 UTC