Date: 11 Sep 1997 23:39:00 +0200
From: kai@khms.westfalen.de (Kai Henningsen)
To: drums@cs.utk.edu
Message-ID: <6deSJW5UcsB@khms.westfalen.de>
In-Reply-To: <19970911165156.5685.qmail@koobera.math.uic.edu>
Subject: Re: Message-IDs and tracing
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Organization: Organisation? Me?! Are you kidding?
References: <19970911165156.5685.qmail@koobera.math.uic.edu>
Comment: Unsolicited commercial mail will incur an US$100 handling fee per received mail.

djb@koobera.math.uic.edu (D. J. Bernstein)  wrote on 11.09.97 in <19970911165156.5685.qmail@koobera.math.uic.edu>:

> > Actually, I was thinking of a fairly successful case of *backward* tracing
> > when I wrote that.
>
> Modern MTAs put all useful information into the message header. The logs
> are irrelevant if you have the message.

Looking at the message won't tell you which of these informations are  
actually reliable, which was an issue in this case. (In fact, it was  
fairly obvious that some information in there was, umm, contrafactual.)

> > With a message id, we can just grep.
>
> With a receipt, we can just grep.

No.

> The difference is that the receipt
> uniquely identifies a delivery, while the same Message-ID is often
> applied to thousands of deliveries.

You're repeating yourself. I already answered that.

> > Actually, I think we just showed why it is very useful.
>
> So far, there are zero examples supporting this claim.

If you think so, we will just have to agree to diasgree.


MfG Kai