RE: Web-friendly SOAP

 Eric,
 I believe SOAP is logically layered below HTTP because SOAP is
not an application protocol. My WebP would be layered on the same
layer as HTTP, SOAP is somewhere between TCP and HTTP - it
provides message + metadata structure and some useful mechanisms
(like mustUnderstand) but it doesn't provide application
semantics.
 Best regards,

                   Jacek Kopecky

                   Senior Architect, Systinet Corporation
                   http://www.systinet.com/



On Sun, 16 Jun 2002, Eric Newcomer wrote:

 > So would this be similar to an earlier proposal, I think from Larry
 > Masinter, to define SOAP as essentially an equal to HTTP?  Meaning at an
 > equal layer, with its own protocol type recognized by the Internet?
 > 
 > Eric
 > 
 > -----Original Message-----
 > From: xml-dist-app-request@w3.org [mailto:xml-dist-app-request@w3.org]On
 > Behalf Of Jacek Kopecky
 > Sent: Sunday, June 16, 2002 12:35 PM
 > To: Eric Newcomer
 > Cc: 'Paul Prescod'; 'Mark Baker'; xml-dist-app@w3.org
 > Subject: RE: Web-friendly SOAP
 > 
 > 
 > 
 >  Eric,
 >  I was suggesting dropping the HTTP binding because HTTP is an
 > application protocol already supporting metadata and I don't see
 > a real value of SOAP as the content type in HTTP. If the HTTP
 > binding stays (which I assume will be the case) and if it is the
 > secondary binding as I'm suggesting, I wouldn't want any
 > duplicity of methods and headers, I'd still put headers in only
 > one place (in an HTTP header or in a SOAP header) and I wouldn't
 > put the explicit method element in the body.
 >  The example SOAP envelope in my email is supposed to be sent
 > over TCP or other similar binding.
 > 
 >                    Jacek Kopecky
 > 
 >                    Senior Architect, Systinet Corporation
 >                    http://www.systinet.com/
 > 
 > 
 > 
 > On Sun, 16 Jun 2002, Eric Newcomer wrote:
 > 
 >  > If I understand this correctly, you are proposing a kind of dual
 > approach,
 >  > for example including the method name in the HTTP binding and also within
 >  > the message itself?
 >  >
 >  > My big concern with the move toward HTTP specific binding for SOAP is
 > losing
 >  > the ability to easily map to JMS, MQ Series, CORBA, J2EE, etc.  I know
 > Mark
 >  > Baker says we can do it anyway, and perhaps we can, but to me one of the
 > big
 >  > questions is whether or not we are thinking about just sending the entire
 >  > message over the transport, and letting the "endpoints" decide what to do
 >  > with it, and how to interpret it.
 >  >
 >  > So you are suggesting including the header information in both the HTTP
 > part
 >  > of the message and also in the SOAP part of the message?
 >  >
 >  > Eric
 >  >
 > 
 > 
 >

Received on Sunday, 16 June 2002 12:59:27 UTC