>At 6:55 PM 7/11/94, Peter Williams
> <"/I=P/S=Williams/OU=cs/O=ucl/PRMD=UK.AC/ wrote:
>>That the Sender: and Reply-To: and From: fields are potentially,
>>logically different?
>
>Definitely. E.g., the Sender field in messages from you reflects the
>machine and account you are sending from. The From field X.400 address you
>use represents an address to which one cannot send mail, since Sprint just
>informed me that they won't relay mail from the Internet. So, you could
>also use a Reply-To field for the ACTUAL address to which replies should be
>sent.
(1) Well the relaying statment is not quite true, is it, Dave. What Merit said
was:
From: Craig Labovitz <labovit(_at_)merit(_dot_)edu>
To: dcrocker(_at_)mordor(_dot_)stanford(_dot_)edu (Dave Crocker)
The Sprint Internet gateway only accepts traffic for Sprint
customers and affiliates. You will have to use another gateway
to reach ADMD=gold.
Disappointing that though delivered, it wasn't relayed, but there we
are: security and money always separate people who wish to communicate.
So, its my fault. Please dont criticize Sprint or British Telecom. They did
what they agreed to do, and its my fault for expecting more. I'll
substitute a formal Euro relay service next time, for RARE-MHS -
MHS-RELAY. They operate service especially designed for researchers
who experiment.
(2) the cited From: address is not an X.400 address; it is an ARPA Internet
address, of the form "Full Name"@domain. Not so different to the
usenet-style "q!t!o(_at_)psi(_dot_)com".
"DCrocker at UDel-Relay: loses; [USER] 550 Unknown domain 'UDel-Relay'" from
the RFC 822 standard title-page, for example of this addressing.
Our need in all this to obtain properly qualified and certified
information about the identity of the communicants, and the signer.
Legal acceptance of signed message content depends on this. If the
Internet truely goes down the use of mail addresses in certificates,
then we have to adapt the current naming pratices somehow along these
lines.