Great Circle Associates List-Managers
(May 1993)

Indexed By Date: [Previous] [Next] Indexed By Thread: [Previous] [Next]

Subject: Re: X-Listserver (Re: Pegasus X-Pmrqc header)
From: Peter Svanberg <psv @ nada . kth . se>
Date: Thu, 27 May 1993 11:29:22 +0200
To: list-managers @ GreatCircle . COM
Cc: psv @ nada . kth . se
In-reply-to: Your message of "Wed, 19 May 1993 13:56:49 +0200." <9305191156 . AA28336 @ ikki>

Quoting John Martin:

> >X-Listserver: Your list server.
>I also agreed to contact this list to try to achieve a consenus for
>identifying mail which has come from a mailing list.  David will try
>to make provision for this in future versions of Pegasus if we can
>(Dictator hat on)
>I think it would be nice if we all agreed on the format of a header to
>be added by our various list servers - would this be acceptable to the
>group? I prefer the 'X-Listserver:' form but I'm open to suggestions.

Quoting:  Stephen R. van den Berg

> I already use a different field.  It was suggested previously (about
> four months ago).
> To every outgoing mail I add:
> X-Mailing-List: <procmail @
 informatik .
 rwth-aachen .
 de> archive/latest/196
> Precedence: list
> The format for the X-Mailing-List field would be:
> 	<submission_address_for_the_list> archive_retrieval_key
> The "Predence: list" field should be added by all mailinglist generated
> mail.  It solves problems with the configuration of old sendmails
> (which don't send bounces on "Precedence: bulk").  And it is supported
> by Sendmail-6.XX and BSD4.4-vacation.

1) I haven't seen any definition of the purpose of the field we
   are trying to agree upon. If it is to identify that "This
   is a message sent out to this mailing list", then
   (X-)Mailing-List is natural. If the purpose is to identify
   which mailing list software was used to distribute the
   message, then X-Listmanager is natural.

2) I think an "archive_retrieval_key" is too much, a
   "sequence_number" is enough. Amongst others, the key can
   become out of date. (In the above example: For how long is
   this list regarded as "latest"?) Then it's more important to
   use the space to inform the user about the name of the

   I suggested in nov -92 (and still do):

7) Add a field on this form:

   X-Mailing-List: [list name/description]*[sequence number] <[list address]>


   X-Mailing-List: Mailing list admin*207 <List-Managers @
 GreatCircle .

   This is something we would like to promote. It is useful for (at
   least) three purposes:

   a) It identifies the message as a mailing list ditto and makes it
      easy for an incoming-mail-sorter.

   b) It serves as a loop detector - any incoming message containing
      this field with the correct list address can be dumped.

   c) The sequence number faciliates for a subscriber to find out if
      he missed some message on the list. Also, it is easier for
      anyone to refer to a number than to a long message-id.

3) If we can come to an agreement on this field, maybe we should
   try to officially register it, as stated in RFC 822? I have
   been told that nobody have done this before so it's about time...
   (If we succeed we could skip the "X-" prefix!)
Peter Svanberg, NADA, KTH		    Email: psv @
 nada .
 kth .
Dept of Num An & CS,
Royal Inst of Tech			    Phone: +46 8 790 71 40
S-100 44  Stockholm, SWEDEN		    Fax:   +46 8 790 09 30

Indexed By Date Previous: Re: Marc . Rinfret @ eng . canadair . ca suggested that I forward this
From: "F. Scott Ophof" <Ophof @ CS . UWindsor . Ca>
Next: Re: X-Listserver (Re: Pegasus X-Pmrqc header)
From: johnw @ bahainvs . org (John Wiegley)
Indexed By Thread Previous: X-Listserver (Re: Pegasus X-Pmrqc header)
From: berg @ pool . Informatik . RWTH-Aachen . DE (Stephen R. van den Berg)
Next: Re: X-Listserver (Re: Pegasus X-Pmrqc header)
From: chip @ chinacat . unicom . com (Chip Rosenthal)

Search Internet Search