ietf
[Top] [All Lists]

Re: Eliminating Virus Spam

2001-01-03 19:00:02
From: hardie(_at_)equinix(_dot_)com

Please point to an example of a useful multipart message seen in
this list or that might someday be useful in this mailing list.

Example:

A multipart/signed message containing a pgp signature known to
some/all members of this list.  As, for example, the ietf registrar pgp
signature (see http://www.ietf.org/ietf-registrar.asc).

I'm not sure that is an example of a useful multipart message in this
or other plausible open international mailing lists. 

  - Messages from the IESG or other parts of the IETF bureaucracy that
     should reasonable be signed seems unlikely to be sent to this list.

  - I don't recall seeing that signature block in this list, and since
     I use a primitive MUA with very primitive MIME support normally
     turned off, I think I would have noticed.
     However, now that you remind me of the IESG's traffic, I think much
     of the IETF announcement list is multi-part.  Strangely, especially
     after this morning's burst, I don't have a sample handy to check.

  - Semi-private messages should not be sent to public, international lists.

  - Because signed messages are rare in public, signing is not much
     good in public lists.  If signed messages did become common that
     would change.  (There are reasons why I doubt they will become common,
     but let's not argue that now.)

  - http://www.ietf.org/ietf-registrar.asc is an awful lot of bits.
     That's not a consideration in some quarters, but few if any
     that I respect.


However, assume for the sake of argument that signature blocks are
reasonble.  Can you suggest a sendmail pattern that would exclude
other than signature blocks?  Perhaps the following?

     HContent-Type: $>+Check_CT
     SCheck_CT
     R$*multipart$*application/pgp-signature$*  $@ OK
     R$*multipart$*             $#error $: 553 reject multi-part junk


Vernon Schryver    vjs(_at_)rhyolite(_dot_)com



<Prev in Thread] Current Thread [Next in Thread>