Re: [EXIM] multipart/report delivery status messages

Página Principal
Apagar esta mensagem
Responder a esta mensagem
Autor: Microwave Systems Eximlist Exploder
Data:  
Para: Georg v. Zezschwitz
CC: Mike Richardson, exim-users
Assunto: Re: [EXIM] multipart/report delivery status messages


Actually the SENDER (or perhaps the sender's MUA) of a message should
be making the decision about what type of bounce it wants. There was an
idea somewhere about tuning bounces to wether or not the original
message had MIME headers.. that might be a step in the right
direction.. Except some people who use MIME-aware MUA's might still
prefer a text bounce..

As far as postmasters being copied on bounces, with the exception of
CONFIG errors, Uhm.. WHY.. Why does a postmaster need to know some dork
mistyped an email address or sent to an account that has been closed?


On Mon, 28 Jun 1999, Georg v. Zezschwitz wrote:

> Hi,
>
> > Most 'people' don't tend to be sys admin who have to deal with
> > the bounces and other mail problems. Mail admin would normally be
> > unix admin who don't have the time to mess with graphical toy email
> > clients.
>
> What I was speaking of is the use of Exim as central mailserver.
>
> The bounces will be received by you, the postmaster, but by the
> sender as well - mostly a poor Windows user.
>
> And if anybody resends a mail, that's hopefully the sender and
> not the postmaster.
>
> > Plain text has worked for a long time, still works and is likely to
> > work for a long time to come. Lets not move just because we can.
>
> uuencode has worked for a long time, does still work, and
> I tend to prefer it for "Unix to Unix"-sysadm-communication :-).
>
> However, there are good reasons why MIME has become a bit more
> popular for attachments than uuencode, and these reasons are
> pretty similar to the reasons, why some people seem to want
> MIME-stylish bounce mails:
>
> - A well defined structure
> - A formal part describing the failure reason, defined by RFCs
> - Content saving
>
> The lack of comfort of a very certain MUA is not a good argument
> against an option for a MTA.
>
> And the old-stylish bounce message does certainly not work fine
> for anything else than plain 7bit ascii messages, and has never
> done.
>
> Greetings,
>
>
> Georg
>
>
> --
> *** Exim information can be found at http://www.exim.org/ ***
>
>



--
*** Exim information can be found at http://www.exim.org/ ***