Received: from localhost by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id AAA18302; Tue, 19 Sep 1995 00:58:21 -0400 X-Resent-To: drums@CS.UTK.EDU ; Tue, 19 Sep 1995 00:58:19 EDT Errors-to: owner-drums@CS.UTK.EDU Received: from dogie.macc.wisc.edu by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id AAA18293; Tue, 19 Sep 1995 00:58:18 -0400 Received: by dogie.macc.wisc.edu; id AA19309; 5.57/42; Mon, 18 Sep 95 23:58:15 -0500 Date: Mon, 18 Sep 95 23:58:15 -0500 From: Eric Norman Reply-To: drums%CS.UTK.EDU@CS.UTK.EDU Message-Id: <9509190458.AA19309@dogie.macc.wisc.edu> To: drums@CS.UTK.EDU Subject: Re: one use of Resent-* > If you autoforward your mail on a VMS system running PMDF > (using "set forward ..." in VMS MAIL), PMDF appears to add > the following headers: > Resent-Date: (the date at which the mail was forwarded) > Resent-From: (the original from address...NOT that of the user > forwarding the mail) > Resent-To: (the address to which the mail is being forwarded) > Resent-message-id: (a new message id for the forwarded message) > I'm not interested in whether this is right or wrong; only > in what the different UA and MTA behaviors are. One of the main mail MTAs that we use here at Wisconsin is a version of PMDF that I modified. It takes the view that addition of Resent-* headers should only happen when there is *explicit* action by a human being to forward (i.e. resend) a message to a different person. That is, Resent-* headers are intended to indicate cases like "Oops, Gwendolyn handles this now or she knows more about this than I do", I'll send this to her. With this view, normal replies (you get only one in VMS) go to the original Reply-To: or From: header, not to any Resent-* header that might have added. The information in any Resent-* headers is for Gwendolyn's use to correct addresses that correspondents should be using to communicate with her. The only "justification" of this view that I can find in RFC822 is the fact that the introduction of Resent-* headers (4.2 FORWARDING) uses the phrase "forwarded by an original recipient who attached"; this means explicit human intervention to me. (Keith, stop reading; you said you weren't interested). In light of the above, I also think that adding Resent-* headers when a message is automatically forwarded is wrong. It's unnecessary clutter that provides no useful information farther on down the line. I know where my .forward files point and the Received: headers will verify that the message went through those hosts if I care. -- Eric Norman