Re: Conneg with intermediaries

Tamura-san,

The first response to your question is:

I wanted to ask the working group for reactions to the basic idea. If
the working group agrees to using this approach, then we can specify
the technical details and the details about its "status", such as
whether it is mandatory or optional.

However I will take your positive reaction as an excuse for me to
suggest specific answers to your questions:


Monday, November 18, 2002, 1:45:26 AM, you wrote:
Hiroshi> In the latest I-D (-03), senarios for relay is described in Appendix.
Hiroshi> If my understanding for your intension is correct,
Hiroshi> the relay case is informative and the description does not
Hiroshi> mandate the behaviour.

I think that it will be better to make this mechanism mandatory to
implement.

It is the only way that an SMTP relay can know a) that the sender is
authorizing transformations, and b) which specific transformations are
authorized.

Of course, this mechanism cannot prevent an SMTP relay from ignoring the
information provided by the sender. But it can ensure that the SMTP
relay has the information.


Hiroshi> Do you think the idea below should be included in Appendix?

I suggest that it be part of the main specification, rather than in
the Appendix.


Hiroshi> Also, although you tell not to mention it now,
Hiroshi> how is the list conveyed?

I have not thought about the details, yet.  I am hoping that we can
use an existing mechanism, but do not know what is practical.

Hiroshi> Regards,
Hiroshi> --
Hiroshi> Hiroshi Tamura, Ricoh Company, LTD.
Hiroshi> E-mail: [email protected]



d/
-- 
 Dave <mailto:[email protected]>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850