Spam Control Considered Harmful

Scott Hazen Mueller zorch at
Wed Oct 29 20:55:20 UTC 1997

>From: Phil Lawlor <phil at>

Welcome to the discussion, Phil.

>>a large-scale re-structuring of Internet mail to provide for secure
>>authentication and cost sharing for received e-mail.

>What if the equivalent of "caller ID" was built into sendmail?  Making sure
>that the sender is a valid email address.

It's a necessary precondition, but not sufficient by itself.  Also, simple
address verification may or may not be enough.  There is no statute or case
law that makes the owner of an address legally liable for the mail emitting
from there - this could be an issue for claims of forgery and the like.

The above notwithstanding, assume for the sake of argument that one could
develop and deploy a secure mail system that authenticated message origin to
the account level.  This would meet the first requirement, and could
*possibly* be the infrastructure for building the second.  However, limiting
anonymity likely wouldn't provide a strong deterrent by itself, since spammers
could still run through multiple non-anonymous dialup accounts over the
lifetime of a spam campaign.

The other piece that would be needed - in this technological solution world -
would be a fast, secure universal electronic payment system such that each
mail message would be accompanied by some offer of payment.  A friend or some
other previous acquaintance would presumably send messages with a 0 cent
payment and assume that you would recognize their name and address and accept
the message anyway.  An advertiser would include some amount and you - or your
filters - would decide to accept or reject the message based on that and
possibly some other tagging information.

The scheme has generally not been sketched in much further detail because the
deployment issues typically overwhelm any discussion.

Scott Hazen Mueller              zorch at          +1 415 281 6550 x269
      Vice President Engineering, Whole Earth Networks (Hooked and The Well)

More information about the NANOG mailing list