On November 20, 2008 at 08:24 schaefer(_at_)brasslantern(_dot_)com (Bart
Schaefer) wrote:
In case you can't tell, the questions I want someone (perhaps Barry) to
answer are: What have I got wrong about this economic model? And if
I've imagined it correctly, can you argue convincingly that it's enough
better than the current situation to be worth the costs of bringing it
into existence?
Well, spammers right now send on the order of a billion messages per
day, that might even be per-each-spammer for the worst offenders.
Their return on that O(billion) messages is presumed to be pretty
small. Maybe thousands of dollars per day in the most lucrative
schemes, and even that only for short spurts. Probably hundreds of
dollars per day average, maybe even less.
But with practically zero costs and zero entry cost currently that's a
pretty good living for some, particularly in some countries.
THEREFORE, the assumptions go, you wouldn't have to raise the cost
very much, either entry cost or per message, to make what they do
unprofitable.
And, since practically nobody else sends anything like O(billion)
messages per day such costs wouldn't bother maybe 99%+ legitimate
senders.
OTOH, I believe that if we could magically set a cost it should also
impose a cost on even quite legitimate senders if for no other reason
than to pay for enforcement of those costs on spammers (perhaps
through something like certificate authorities which sell SSL certs
now, as one structure.)
That aside let's try an example.
Say we could effectively impose a 1c/million messages cost (I'm using
US dollars because I might need to buy raw petroleum with the income
stream :-)
And, further, call any charge in a 24 hour period of less than 1c
equal to zero.
So, the O(billion) spammer would be in for $1,000/day.
Assuming we could impose that with some reliability, let's not go
running there yet, how many of the spammers that we actually worry
about could stay in business if they had to pay $1,000/day?
Maybe a few. That's probably too low.
One hazard here is that we're also potentially removing some of their
costs, or at least making their campaigns potentially more effective
if we say pay the postage and you're ok. Individuals might still
choose to block you as much as people throw out paper junk mail unread
today, but you'd be out of the blacklists etc if you're playing
fairly.
So make it 1c/100,000, that's $10K/day to the spammer, and still less
than 1c for most normal users, even large public mailing lists. A huge
public list might be 1M msgs/day, still only $.10/day, $36.50/year,
and there could be a way to waive even that.
Someone like Amazon maybe sends out a few million messages per day, I
have no idea, let's call it 20M/day. That would be $2/day. That still
seems too low but you get the idea. Maybe Amazon et al should pay
O($1,000)/day, the spammer should face O($100,000/day), and the
average user 1c approaching zero.
Ok, a huge list at that point might be a few hundred dollars per year,
not that unthinkable, maybe the list should pay a few hundred dollars
per year much like anyone has to pay for an SSL cert. Anyhow, it's
well within the range for a waiver or special rate for non-commercial
use or whatever, it wouldn't take much of a discount to make $300/year
palatable.
Anyhow, that's what I'm thinking as a goal framework.
Why would someone like Amazon agree to be charged say $300K/year?
They would if they believed it drastically lowered the spam, suddenly
their email might get read instead of drowned out in the deluge.
Obviously how to get there is non-trivial. But if it *were* trivial
why are we all still here?!
--
-Barry Shein
The World | bzs(_at_)TheWorld(_dot_)com |
http://www.TheWorld.com
Purveyors to the Trade | Voice: 800-THE-WRLD | Login: Nationwide
Software Tool & Die | Public Access Internet | SINCE 1989 *oo*
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg