ietf-asrg
[Top] [All Lists]

Re: [Asrg] Re: What are your criteria for the end of spam?

2003-06-06 16:02:46
Rodney Tillotson wrote:
At 26/05/2003 07:02 +0900, Shannon Jacobs wrote:

A few hundred messages later (on 7 June)... First, a meta-observation. As
I've already remarked, the volume of traffic here reminds me of the volume
of spam one of my old spamified email accounts received. Yes, there is an
actual signal here, but there's also a lot of noise and fury.

Observation 1: The sheer volume of ASRG email makes it essentially
impossible for people without a whole lot of time to participate
effectively.

Conclusion 1: The discussions tend to be dominated by a few people with lots
of time. Maybe they'll cover all the bases, but I doubt it. In particular,
those few people are apparently uninterested in economics, and I remain
convinced that the heart of the spam problem is economic. As long as the
spammers can dream up new get-rich-quick schemes, they will continue trying
to devise new ways to break any system. (However, I do like several aspects
of the RMX and TitanKey C/R systems.)

Observation 2: The essential independence of email messages results in lots
of quoting.

Conclusion 2: That increases the redundancy, increases the volume, and makes
the situation worse, and suggests that email is a poor tool for this job.

Now I'm going to flog those dead old economic issues again...

1) We want to be informed of the best values.
2) We don't want to give up our privacy.
3) Companies compete to create the best values.
4) Email could help customers find the best values.

If (1,4) _I_ want to be informed, _I_ go and do something about
it. I opt IN to a few things (even the ASRG list ...). I don't
want to be informed of stuff on the basis of someone else's
values.

Very good point, which I why I think the recipient should set the values and
the advertisers should pay up or go away. However, on additional reflection,
the problem seems very intractable because my value for information changes
all the time. Concrete example with "me" as the email recipient and "you" as
the advertiser with a car to sell:

When I am not interested in buying a new car, I do not want to receive car
ads, and I'd want to charge you a lot of money if you insist on my seeing
your ad. In that state of mind, I'm also a bad prospect, and so you would be
wasting your money to pay for me to see the ad. The obvious conclusion is
that you won't bother me, but will spend your advertising money to reach a
more likely prospect.

It's actually pretty simple statistics. You compare the negative and
positive expectations. The cost of sending email is a negative expectation,
while the positive expectation is the profit from the sale times the (small)
probability that I will actually buy it from you. Arbitrary example, but
suppose I charge you $10 to read your email, the profit from the sale is
$500, and the probability for my type of customer is 1%. You'd have to email
100 such customers to make one $500 sale, but the cost would be $1,000
(100x$10), so rationally, you won't do it. (Of course the spammers think of
0 as the cost, so any positive expectation, no matter how small, apparently
justifies their spamming.)

However, if I actually am interested in buying a new car, the situation is
very different, and I actually would highly value receiving your email. It
might save me a lot of money if you really are offering the best deal. Even
if you aren't offering the best deal, you would be providing data that would
help me recognize the best deal. If I've moved into a category with a 10%
probability of buying, then it is very much worth it for you to pay the $10
to reach me.

Much more to say on the economic topic, but that's all the time I can spare
from my actual life...

_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg



<Prev in Thread] Current Thread [Next in Thread>