Re: A potential way to cut down on spam.

[Eugene Crosser <crosser@rol.ru>]

On Wed, 2004-02-04 at 18:24, alvin wrote:

> >>We are going to run this up at some point and give it a try but it may 
> >>take a bit of time before we can find the time. So I thought I would 
> >>post the suggestion and see if anybody has any comments or has tried 
> >>this before.

> >We did not try this because we understand that this will only work until
> >spammers undestand what happens and adjust their techniques.  (For us,
> >this may happen rather quickly as we a "target of choice").

> for spamming to work it requires high volumes and cannot tolerate 
> retries. One of the projects here(not mine or sanctiond by me) is a 
> service that mails out jokes and other humorous spam to people who opt 
> in. This works by sending out about 500,000 mails a day in a 2-3 hour 
> period the system will only work because most of the addresses take mail 
> the first time and failures are taken out of the high speed queue and 
> put in a regular mailer queue.
> The regular mail takes as long as it needs to and failures are removed 
> from the mail list after a while. This scheme works only because there 
> is about a 99 to 1 ratio in the directly deliverable mail to mail that 
> requires special handling.
> 
> One of the up sides of spam is that I can send out 1,000,000 mails in a 
> day and about 1% will be read. But if I am now limited to 10,000 mails a 
> day because I cannot get through first time then spamming is slightly 
> less attractive.

A single system that sends 500,000 spam messages a day is not a problem
at all to deal with.  Most of our incoming spam is from thousands of
infected "zombie" home PCs, each of them trying  to send a mere hundred
messages.  For spammers, it won't be any problem to program those to
make several delivery attempts.

Eugene

-
To unsubscribe from this list: send the line "unsubscribe zmailer" in
the body of a message to majordomo@nic.funet.fi