From: "Eric S. Johansson" <esj(_at_)harvee(_dot_)org>
...
I go to a web site and purchase something. They send me an email
receipt. How exactly are they going to automatically get through my
challenge response? ...
problem has been solved with stamps. I know I'm sounding like a broken record
(or will very shortly). The model is if you filter for stamps, then a robot
can
attach a stamp to an auto response and it will get to your inbox. ...
No, that is not a solution, because it is based on the engineering of
wishful thinking. So is absolutely any other idea related to the
Internet that begins with "Install X at most Internet X" when X is a
kind of hardware or software operated by 1,000 or more people.
Anything that requires actions by 1000 people or 0.01% of the relvant
population, whichever is smaller, before it is useful will probably
never happen and certainly cannot be relied on.
Unless the first few dozen individuals who install your proposed change
benefit significantly, there will never be another few dozen individuals
who install it. People do not do things unless they see some real
benefit from it or progress toward their goals.
Charging for email is an exceptionally hopeless idea, because it starts
with "first a large minority of users install email stamp software
and buy stamps." Until at least 10,000,000 people regularly stamp
their messages and require stamps on incoming mail, stamps are only
overhead that will merely cause legitimate mail to be lost.
If the next XP automatic update installed the necessary software, you
would still not have critical mass. You would still need to convince
the first 10,000,000 to buy e-stamps. Until all of your correspondents
had bought e-stamps, your own roll of e-stamps would be useless. Your
friends would see how little you got out of the exercise, and somehow
never get around to buying their own rolls of e-stamps...or signing up
with Verisign for a PKI email non-spamming identity or any other
proposal that starts "first you get everyone to ..."
When do you think the IESG would install e-stamp software on all of
the IETF WG mailing list systems? Before you answer, note that most
of the spam I suffer comes from IETF lists because their operators
refuse to install minimal spam filtering such as rejecting mail entirely
in Asian character sets.
Look at history:
- PayPal looks like it could survive, because the first people who used
it got some good out of it. If you need to pay or receive $10,
rendezvousing on a PayPal account is easier than sending a check.
- the WWW took off because HTML was somewhat useful for the first
users as a mark-up language and a very few individuals at enough
big outfits (e.g. CERN) pushed it as a network thing that it
was immediately useful to others at those big outfits and eventually
outside.
- Paul Vixie's RBL took off because the first people who used it
it benefited. It was similar for all other well known blacklists
and non-distributed filters.
- the DCC took off and will survive as long as new versions of the
checksums stay ahead of the spammers, because the handful of
outfits that first tried the DCC had enough mail traffic to exceed
its required critical mass and so got some spam filtered that
would otherwise have been delivered.
On the the other side, the dot-com boom and bust is filled with ideas
that went "first we'll get lots of users, and then those users will
find it useful and we'll get more users." Look at the history of
"micropayments." That idea is quite similar to e-stamps, and was a
dud. If you don't remember microspayments see
http://www.openp2p.com/pub/a/p2p/2000/12/19/micropayments.html
In the real world, if the first "early adopter" users don't find it
useful (whatever it is), you will never have a second batch of users,
and the early adopters will soon drop it.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg