Micropayments, redux

[Pete Chown]

David Wagner wrote:

> Yes, but the probability of it being significantly worse than I claimed
> (i.e., by more than a factor t) is exponentially small (in t).  One can
> easily calculate concretely exactly what the risk curve looks like.
> I'll spare everyone the details and just say that I see no reason why
> this should be a showstopper in practice.

Actually I think it may be a showstopper in practice for rather 
different reasons -- people's behaviour when exposed to risks is rather 
odd.  The British lottery, for example, pays on average £0.50 for each 
£1 ticket.  No one would buy a 50p piece for £1, but people will buy an 
expected win of 50p for £1.  People worry more about their children 
being the victim of a paedophile than getting run down by a car.  People 
worry more about the tiny risk from the measles vaccine (if it is a risk 
at all) than about the risk of dying in a measles epidemic.

I don't know which way the psychology would work in this case.  Would 
people be put off buying content by the small chance of paying a high 
price?  Alternatively, would they enjoy the gamble, and the fact that 
most of the time they would get content free?  I think it may be the 
former, because the protocol is like the lottery in reverse.  The 
lottery promises you a small chance of a big win.  This micropayment 
scheme gives you the small chance of a relatively big loss.

There is also the risk that the system could be classed as an illegal 
lottery in some jurisdictions.

-- 
Pete


---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at wasabisystems.com