On Sun, 10 Feb 2002, Karl J. Runge wrote:
> I naively thought Errors-To: was the only way to do this. (I grepped and
> saw this string in the majordomo code, but that was likely regarding
> subscription, etc, activities, not posts)

  A lot of software support(s|ed) the use of "Errors-To", because at one
time (when UUCP was the norm and SMTP was unusual), it was the only way to
accomplish what we are talking about.  I suspect that is what you saw.

> Let me see if I understand you, a posting's header might look like:
> 
> >From [EMAIL PROTECTED]
> Date:
> ...
> From: Benjamin Scott <[EMAIL PROTECTED]>
> To: "Greater NH Linux Users' Group" <[EMAIL PROTECTED]>
> ...
> 
> So replies would (likely) go to the "From:" address, but errors to
> the "From" address?

  Incorrect.  That inital "From " line is part of the message storage
mechanism (your Unix mailbox), and has nothing to do with message transfer.  
The actual message begins with the line immediately following the "From "
line.  (Note the distinction between the "From " line and the "From:" line.  
The former is part of the message storage mechanism; the later is part of
the message itself.)

  (What follows is a brief (!) explaination.  Even so, I have to go into
some detail.  Casual users will want to stop reading now, to avoid
eye-glazing problems.)

  Internet messages, at the highest level, consist of an envelope and the
message contents.  The content is the entire message you see on your system.  
The envelope is the conceptual entitity used to transport the contents
between systems.  Users never see the envelope.

  In more pragmatic terms, the contents of the message begin with the first
message header, and end with the last line of the message body.  The
envelope is the SMTP session that was used to transfer it between systems.

  When an SMTP message transfer is performed, I (as the sender-SMTP) provide
a reverse-path (MAIL FROM) and one or more forward-paths (RCPT TO).  The
forward-paths specify which mailboxes the message should be delivered to.
The reverse-path is used if a DSN (Delivery Status Notification, AKA "bounce
message") needs to be sent back, e.g., because a forward-path proves to be
undeliverable.

  When you (as the User Agent) read a message, you see only the message
contents.  If you wish to reply to the message, you look for a "Sender:" or
"From:" header, and use the listed address.  You have no knowledge of the
envelope.

  Typically, the reverse-path and the "From:" header corespond.  Likewise,
the forward-path and the "To:", "Cc:", and "Bcc:" headers usually add up to
the same set.  They can, however, differ.

  In the case of a mailing list like this one, the typical behavior is: The
"From:" header indicates the original sender of the message (e.g., you or
me), while the reverse-path is the address of the list supervisor (which
might be a person or a robot).

  (The above, of course, contains simplificiations and omissions.  You can
find the full details in RFC-2822 (Internet Message Format) and RFC-2821
(Simple Mail Transfer Protocol).  Those two documents define what we think
of as "Internet email".  They are the new versions of the classic RFC-821
and RFC-822 documents, accurately described as "the most popular RFCs
ever".)

  Are you sorry you asked, now?  :-)

-- 
Ben Scott <[EMAIL PROTECTED]>
| The opinions expressed in this message are those of the author and do not |
| necessarily represent the views or policy of any other person, entity or  |
| organization.  All information is provided without warranty of any kind.  |




*****************************************************************
To unsubscribe from this list, send mail to [EMAIL PROTECTED]
with the text 'unsubscribe gnhlug' in the message body.
*****************************************************************

Reply via email to