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

Re: SOAP Binding Framework Concerns

From: Marwan Sabbouh <ms@mitre.org>
Date: Mon, 22 Oct 2001 10:37:34 -0400
Message-ID: <3BD42F2E.DA43B4C@mitre.org>
To: Noah_Mendelsohn@lotus.com
CC: skw@hplb.hpl.hp.com, xml-dist-app@w3.org
Hi Noah;

Please forgive me if I am wrong, but let me see if I understand you.

You wrote:

1) We need to give guidance to those who invent the bindings of particular
transports into SOAP. ( Statement 1 )

Did you mean to say??
We need to give guidance to those who invent the bindings of a particular
feature ( such as Security, reliability, etc..)  into SOAP.   I said that
because you also wrote:

Those rules are invariant across
transports, and therefore should be captured in an invariant manner.
(Statement 2 )

Statement 1 and statement 2 seems at odd.

You wrote:

2) Whether WSDL is used or not, applications benefit from knowing whether
a pattern such as "Request/Repsonse" has the same behavior over two or
more transports, or not.  If it does, then the application can (if
desired) be written to be largely independent of transport.

Absolutely, This should   ALWAYS be the case.

You wrote:
but the binding framework
and MEP framework set out what's comment to all such patterns, and allow
one to make rigorous statements about which patterns are supported by
which transports and bindings.

I think this is outside the scope of this working group.  The best way to
specify that  a particular message pattern is  suitable across different
transports is to define that pattern first, then binds it to those
transports.  WSDL does exactly that.

Thank you
Marwan



Noah_Mendelsohn@lotus.com wrote:

> Marwan:  I think the fundamental reaons the framework is appropriate are:
>
> 1) We need to give guidance to those who invent the bindings of particular
> transports into SOAP.  How do you know if you've done one that is
> conformant?  How do you know if you've delivered the right information to
> the next node at the right time?  Those rules are invariant across
> transports, and therefore should be captured in an invariant manner.
> Regardless of transport, you'll be given an envelope (for example), and
> will be expected to move it to the next node (or fault, or whatever.)
> Those rules are to a large degree common across transports.  The framework
> captures that commonality.
>
> 2) Whether WSDL is used or not, applications benefit from knowing whether
> a pattern such as "Request/Repsonse" has the same behavior over two or
> more transports, or not.  If it does, then the application can (if
> desired) be written to be largely independent of transport.  Indeed, WSDL
> seems a good place to document that commonality, but the binding framework
> and MEP framework set out what's comment to all such patterns, and allow
> one to make rigorous statements about which patterns are supported by
> which transports and bindings.
>
> 3) Similarly for features such as various forms of security, reliable
> delivery etc.
>
> ------------------------------------------------------------------------
> Noah Mendelsohn                                    Voice: 1-617-693-4036
> Lotus Development Corp.                            Fax: 1-617-693-8676
> One Rogers Street
> Cambridge, MA 02142
> ------------------------------------------------------------------------
Received on Monday, 22 October 2001 10:45:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:04 GMT