Fwd: Re: SCMP V.07 to Informational RFC

I've just posted a revised draft of SCMP (it should be up in a few days) 
that deals with the negotiation of payload content between trading partners 
prior to processing requests.

In Patrik's notes, he raises a few additional points (one of which is 
transport negotiation). I'm wondering if this should be handled in the same 
manner as payloads or if I should specify http?  Any thoughts?


>X-Sender: [email protected]
>Date: Thu, 10 Aug 2000 06:56:45 +0200
>To: Tom Arnold <[email protected]>
>From: Patrik Fältström <[email protected]>
>Subject: Re: SCMP V.07 to Informational RFC
>Cc: "Jeffrey I. Schiller" <[email protected]>, Ned Freed <[email protected]>,
>         "Marcus Leech" <[email protected]>
>
>At 11.48 -0700 00-07-31, Tom Arnold wrote:
>>Patrik, this is a resend as I had your old email address... I also sent a 
>>copy of this to Jeff and Marcus, Security Area Directors.
>
>I have checked draft-arnold-scmp, and I don't see that it can be published 
>now either.
>
>The document doesn't specify enough to ensure interoperability between two 
>independently developed applications which each is developed with _only_ 
>the paper in question as a basis.
>
>For example, you don't even define the error codes or transport protocol 
>to be used! How is that to be negotiated?
>
>Secondly, you don't talk at all about how the client and server is to know 
>what SCMP-message-type is to be used. Is there a registration mechanism? 
>What is the review process for message types? Why doesn't you use Media 
>Types? Is there a content-negotiation process for message types?
>
>Why should one use this and not an email message with an EDI containing 
>MIME part in it?
>
>I.e. this is not complete. You now discuss security in the new version, 
>aswell as fixing the reference to S/MIMEv2 which is now gone. Good.
>
>But, more work is needed, much more work.
>
>     Patrik