From: Bill Cole (postfixlists-070913billmail.scconsult.com)
Date: Mon Jan 07 2008 - 09:12:58 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

At 3:49 PM +0100 1/7/08, mouss wrote:
>Bernhard Reiter wrote:
>>There is a method proposed to reduce spam:
>>Simulate a tar pit so that spammers will give up,
>>by stuttering the first n bytes of an SMTP connection.
>>
>
>few comments:
>
>- like some other approaches, this is only effective if a lot of
>sites implement it. otherwise, spammers have a lot fo available
>resources.
>
>- Beating this is not hard. a concurrent client can talk to other
>places instead of waiting for the response.

This is a very important point. Some spammers have already figured
out that slowing tactics exist and have adapted their client
strategies to work with them.

>- you need to know whom to tarpit. if you use your own logs, it may
>be too late (the same client won't necessarily connect again to your
>server). if you use "public" lists, you need to update regularly (I
>know of no list with rsync access, so you're gonna download a full
>list all over again).

The real issue is not whom to tarpit but whom NOT to tarpit and how
to divide that class into senders whom you don't impede because they
are either innocent or immune to tarpitting and those whom you
swiftly reject mail from.

Tarpitting is something you should only be doing to total strangers,
because it is a bad thing to do to your friends and is too kind to
your enemies. :)

--
Bill Cole
billscconsult.com

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]