There is a proposal to start an IRTF research group on the topic of
SPAM. Perhaps one of the things that could be looked at is how mailing
lists could apply spam defenses and still maintain open-ness.
Many of the problems with spam are now the second order effects of
dropped messages that should have been forwarded.
By default SpamAssasin is configured to consult a blacklist that is
currently blocking all UUNET addresses because the maintainer does not
like something that a UUNET subscriber is publishing on their site. I
conclude that blacklists of that type will not last long as the Internet
starts to re-route arround the censorship damage.
That said, there is an old proverb in politics about what you should do
when you become the story. I think the issues that concern the group
members here are more than just the alleged filtering of Bernstein's
From: Keith Moore [mailto:moore(_at_)cs(_dot_)utk(_dot_)edu]
Sent: Friday, January 10, 2003 8:00 PM
To: Thomas Narten
Cc: moore(_at_)cs(_dot_)utk(_dot_)edu; djb(_at_)cr(_dot_)yp(_dot_)to;
Subject: Re: Dan Bernstein's issues about namedroppers list operation
Specifically, all mail sent to namedroppers is:
1) first run through spamassassin. Mail that is rejected here is not
archived, as the number of such messages is large. All
mail sent to
mailing lists on the server hosting namedroppers is run though
spamassassin, so this is not a namedroppers-specific procedure.
SpamAssissin needs to be shot. Most of its criteria are
really poorly chosen.
Even if its criteria are good at identifying spam on a large
scale (with few
false positives) that doesn't mean they will work well for a
discussion. In my experience SpamAssassin has too high a
false positive rate
to be trusted without human review as a backup.
p.s. regarding messages that did not make it to the
namedroppers archive -
are the IETF archives still using to/cc message headers to
archive a message should be stored in? if so, is it
possible that a message
which was sent to multiple lists might be archived in only
one of those lists?
to unsubscribe send a message to
the word 'unsubscribe' in a single line as the message text body.
Description: S/MIME cryptographic signature