headers
Steve Hole | 18 Jul 1997 19:32
Picon

Re: draft-ietf-drums-msg-fmt-02.txt


I took this response off of the DRUMS list because it is along the lines 
of a "me too" response, with a suggestion for SIEVE.

On Sun, 29 Jun 1997 21:42:42 -0400 Pete Resnick <presnick <at> Qualcomm.Com> 
wrote:

> What DRUMS is referring to as "resending" and using "Resent-*" headers 
> for
> I've heard called "resending", "redirecting", and "bouncing". All mail
> packages I've ever heard of use "forward" in the same way that DRUMS does.
> The only place that the word "forwarding" is used to indicate what DRUMS is
> calling "resending" is, unfortunately, in RFC 822. But 822 does not make
> the distinction and does not recognize the act of sending a message as the
> contents of another message. 822's use is simply confusing the issue, as it
> clearly did in the sieve draft.

Making sure that this is consistent between the DRUMS (correct!) 
definition and SIEVE is very important.   User's will regularly use the 
definition that best fits their need of the moment and try to argue the 
point of what a forward truly is.   SIEVE must define it's operation as 
a "resend" operation, not a DRUMS "forward" operation.   The name of the 
operation should reflect exactly what it does.

It seems to me that there could be two operations:

"Resend" - redirect a mail message intact to a new recipient(s) with 
           "Resent-" headers.   This operation is used to automatically 
           redirect a mail message to another recipient without comment.
(Continue reading)

Permalink | Reply | 1 comment |
headers
Tim Showalter | 25 Jul 1997 21:57
Picon

toss action

I'm probably going to change toss to trash in the next revision of the
draft in order to ease some international tensions.  Those wondering why can
contact me.

If you have a better word than trash, please let me know.

--

-- 
                                           Tim Showalter tjs <at> andrew.cmu.edu
Permalink | Reply | 0 comments |
headers
Chris Newman | 28 Jul 1997 18:15

Re: draft-ietf-drums-msg-fmt-02.txt

On Fri, 18 Jul 1997, Steve Hole wrote:
> Making sure that this is consistent between the DRUMS (correct!) 
> definition and SIEVE is very important.   User's will regularly use the 
> definition that best fits their need of the moment and try to argue the 
> point of what a forward truly is.   SIEVE must define it's operation as 
> a "resend" operation, not a DRUMS "forward" operation.   The name of the 
> operation should reflect exactly what it does.

The real problem here is that there are two operations called "forward"
which are distinct from the "resend" operation.

There is an "MTA forward" -- which simply alters the envelope recipients
with no changes to the message.  This is what the Sieve draft is referring
to and this is what a ~/.forward file does on Unix.  Think of this as what
happens when you tell the post office to "forward" your snail mail.

There is an "MUA forward" -- which is the forward operation provided by
MUAs.  This is the one where new recipients are entered, new text is
added, and the old message is enclosed (as message/rfc822).  Think of this
as what happens when you take a message out of the envelope, add a cover
letter and drop it in a new envelope.

Then there is "Resend" -- an MUA action which doesn't alter the message
body, but adds new Resent-* headers and creates a new envelope.

The key point, IMHO, is that MTAs don't add Resent-* headers, and Sieve
is run by an MTA.

		- Chris
(Continue reading)

Permalink | Reply | 0 comments |

Gmane