At 2:23 AM -0500 3/12/03, Damien Morton wrote:
If ISPs charged each other for receiving email, then net producers of
email (and spammers) would end up paying where they pay nothing now. If
Senders pay part of the cost now--it's hidden in their bandwidth
costs. Spammers in general don't pay because they are stealing
bandwidth.
a user at an ISP produced a disproportionate amount of email, the ISP
would have an interest in pushing the cost onto the user, just as users
of excessive bandwidth tend to be charged for it by ISPs.
If ISPs felt that it was profitable to charge for outgoing email
bandwidth they could do it now. Many already block port 25
outbound--so they can easily track outbound email use. They don't
charge for this because there is a perception that users wouldn't
stand for paying extra for it. And also because the cost of sending
email is very low--in the noise compared to most network use. The
cost is on the receiving side, where they have no control.
So I'd argue that an ISP is not going to charge users for sending
email unless someone is charging *them* for sending email.
At the 10% stage, a charging system could act as a filter in which
unpaid emails would be flagged as possible spam. Alternatively, mail
from non-paying sources could be delivered at a lower priority and/or
put through rigorous spam screening with a high false-positive rate.
Paid email always get through.
When important email does not make it through there are two
pressures. One is to on the sender, who wanted the message to get
through. The other is on the recipient, who wanted to receive the
message.
I think it's fair to argue that the importance of getting a message
through to any one person decreases with the number of people you are
sending to. So if I'm sending to twenty people, the fact that one
fails is less important than if I'm sending to one. On the other
hand, the importance of *receiving* a message is based more on the
content than the number of other people who received the message.
To give a concrete example. It's very important to me to get my
order confirmation from Amazon. Amazon, on the other hand, is not
going to be terribly upset if one person in a thousand doesn't get
their order confirmation.
In talking to customers, it's clear to me that most people care *far*
more about false positives (missed email) than false negatives
(received spam). There are exceptions--but that view seems to be far
more prevalent.
Most messages are delivered to more than one person.
This should lead to a simple formulae for the which way the
incentives go when someone adopts a new solution. Unfortunately I've
been up all night and formulae's, simple or otherwise, escape me.
Anyone care to try?
But the point I'm trying to make so awkwardly is that your 10%
solution is does not create an incentive for people to start paying
postage. It creates a very strong incentive for the recipient to
stop charging. That's because the cost of a lost message is far
higher to the recipient than to the average sender.
We've had several people now propose that unpaid/unverified email be
slowed down or filed elsewhere or otherwise demoted in order to
increase the incentive on the sender to follow the rules. I think
those suggestions are actually counter productive. I'd be more
willing, as a recipient, to charge senders postage if I did *nothing*
to the people who didn't pay. But of course if I do that, the
senders will be even less likely to adopt the system.
Fundamentally the tragedy of the commons is not just creating spam,
it's also preventing a solution--because even the non-spammers are
primarily motivated by self-interest.
--
Kee Hinckley
http://www.puremessaging.com/ Junk-Free Email Filtering
http://commons.somewhere.com/buzz/ Writings on Technology and Society
I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg