[Mailman-Developers] Re: Anti-spam "killer app"?

[Michael Meltzer]

 think their are simpler ways here, I would setup a new email address, then I would sent a messages to ALL news group like:

to: alt.sex
from new@address.com
Subject: A new Method:

Hi all my email is new@address.com, please do not use this address, I only want the S P A M Robots to Find it, I am trying out a new
filtering system and need a large collection of S P A M, Moderators please let this though, the archives are a very "fertile ground"
for this guys to harvest address. If you are wondering what I am up to www.picksomething.com/method.html

THANK YOU

BTW make sure the new@address.com is on the webpage. My bet in a week you will have a large sample size for the "spam" side, growing
by the hour, could send it right to the file ;-)be carefull of keywords the harvesters might react to like abuse@foo.com. Thier is
really only one spam table need to get started(the seed) that can be used by all. The personial side mosty likly can be seeded using
the user mail mbox/archives on thier machine(or a list archive)

MJM
PS. I know spaming the news group is a issue here, but minor(they get flooded now), but sometime you have to play by thier rules :-)


----- Original Message -----
From: "David Champion" <dgc@uchicago.edu>
To: "Dale Newfield" <Dale@Newfield.org>
Cc: "Bob Puff@NLE" <bob@nleaudio.com>; <mailman-developers@python.org>
Sent: Thursday, August 22, 2002 1:32 PM
Subject: [Mailman-Developers] Re: Anti-spam "killer app"?


> * On 2002.08.22, in <Pine.OSX.4.44.0208221308140.5976-100000@core.newfield.org>,
> * "Dale Newfield" <Dale@Newfield.org> wrote:
> > On Thu, 22 Aug 2002, David Champion wrote:
> > > "forward to spam-submit@uchicago.edu" is pretty simple, and as the
> > > filter gets smarter, it becomes less necessary to interact with it.
> >
> > That has huge potential risks.  If there's someone I don't like, I could
> > just forward all his mail to that address (it'll learn very quickly if
> > that person has a .sig), and he/she's screwed.
>
> Yes -- I should have mention this in my first message. I'm certainly
> wary of that, but I also feel that sufficient safeguards are possible
> in principle. Verify the sender's envelope address, and require the
> SMTP connection to come from a limited range of addresses. It still can
> be abused locally, but that might be a worthwhile tradeoff, should an
> individual user choose to go that way.
>
>
> > Would there be an equivilant "you have to send all your good mail to this
> > address"?  The privacy risks there are tremendous, as well.
>
> No. I don't want people sending the system lots of legitimate mail. I'll
> specifically ask that people not send legit mail except as necessary; an
> ever-expanding database is a big perfprmance hit.
>
> However, such text would be hashed, and the original discarded, so that
> privacy is protected.
>
> For people using immediate IMAP folders as sources, there are no extra
> privacy concerns incurred.
>
>
> I'm happy to continue talking about this (it's a really interesting
> topic, to me), but maybe it's getting a little off-topic for the
> mailman-developers list. Does anyone have an alternative suggestion?
> Does nobody mind?
>
> --
>  -D. We establised a fine coffee. What everybody can say
>  Sun Project, APC/UCCO TASTY! It's fresh, so-mild, with some special coffee's
>  University of Chicago bitter and sourtaste. "LET'S HAVE SUCH A COFFEE! NOW!"
>  dgc@uchicago.edu Please love CAFE MIAMI. Many thanks.
>
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman-21/listinfo/mailman-developers