On 29 August 1998, Stan Ryckman <stanr(_at_)sunspot(_dot_)tiac(_dot_)net> wrote:
At 09:03 AM 8/29/98 +0300, Liviu Daia wrote:
On 28 August 1998, Stan Ryckman <stanr(_at_)sunspot(_dot_)tiac(_dot_)net>
wrote:
At 07:45 AM 8/28/98 -0700, John D. Hardin wrote:
A suggestion:
If you're going to send a spam-filter snippet to the list, try to
avoid attaching it as plain text. Certain key strings (such as the
"*ull's *ye *old" in the recent example) may trigger filters.
Huh? Just white-list this (and any other subscribed) mailing list
before filtering on subject or content. That's easy and I believe
has been suggested numerous times as a necessity for any reasonable
content-based spam detector.
[...]
You must be joking, this list is the only one through which I
still receive spam.
Sorry, I do not see the purported joke.
Neither do I actually. It was only an attempt to be polite.
While the issue of whether this list should require subscription
to post has been beaten to death (actually, I think requiring
non-subscribers to confirm the post with an OK would do it since most
spam is unreplyable), how much of the spam that made it onto this list
had the list address in the To: or Cc: header?
I don't bother doing this myself, so this is untested, but if you
insist:
[...]
Actually, I did read your message before replying to it, so I
probably understood what you were suggesting from the beginning. :-)
But my problem was not whether to do it, neither how to do it --- I'm
actually already doing something like that. My point was some spammers
_do_ put list's address in the "To:" header. It happened in the past
(in fact, that's what convinced me to add other methods to my anti-spam
filters), and I don't see any reason why it should stop happening
now. Exactly how much spam I get this way is irrelevant, if it can be
stopped, I don't want to see it.
Regards,
Liviu
--
Dr. Liviu Daia e-mail: daia(_at_)stoilow(_dot_)imar(_dot_)ro
Institute of Mathematics web page: http://www.imar.ro/~daia
of the Romanian Academy PGP key: finger
daia(_at_)stoilow(_dot_)imar(_dot_)ro