Fallacious use of 'bank' in net payment systems

[Ian Grigg]

Anton Stiglic wrote:

> Payment systems don't need a PKI, but
> a centralized server as you said, and
> also needs some kind of financial
> institution to bootstrap things (which is easier
> to do in a closed system, than in an open
> system like the Internet).

And, bear wrote:

> (2), Each
> and every consumer system will need configuration and
> a relationship with a "bank" somewhere to cash such
> tokens made to them - and that requires the attention of
> the consumer again to maintain a business relationship
> which he won't bother to do unless there's a significant
> (more than a few dollars a month) revenue stream.

One of the more pervasive fallacies
in the digital payments world derives
from the use of the words 'bank' and
'financial institutions'.

The original inventor of both the
concept of digital cash, and the first
extant system (Dr David Chaum) made
the assumption that the centralised
entity that issued the value was a
traditional banking institution, a.k.a.,
a Bank.

Experience showed this was a bad choice.
Proper business analysis - the sort that
is taught in b-schools - also shows that
a Bank as the issuer is a bad choice.
(I beg indulgence on this point, one
would normally have to show this, but
space & time are not sufficiently elastic).

Sadly, the word 'bank' lives on.  And it
forcefully implies to everyone that the
center should be a Bank.  Unfortunately,
this implication has contributed to the
destruction of all competant cryptographic
payments systems, and cost us countless
VC millions.

Fast forward to the spam situation.  We
have several actors:  sender & recipient,
sender's ISP & recipient's ISP.  Spammer.

It takes only a moment's thought to
realise that if a payments solution is
to be used to address the spam situation,
then the ideal place for the issuer is
inside the ISP.

  * Each ISP already has a billing arrangement
    with the user,
  * Within the ISP, makes for short fast links
    between the mail payments client, the
    issuance server, the mail forwarding
    server, and the user's mail client,
  * Each ISP already has a relationship with
    each other ISP,
  * Each ISP having its own issuer of email
    postal stamps will more or less efficiently
    match to the scaleability issue of email,
    in that it adds O(1) complexity of links.
  * Each ISP already knows who is mailing whom,
    so there is no need to preserve the privacy
    of payments.

Etc, etc.

I personally use the word Issuer for the
central entity in a payments scheme.  In
token money (a.k.a. DBS / blinded coins)
the word Mint is often used.

The key thing here is that the words 'bank'
and 'financial institution' should be
expunged if ever one had intentions of
building a successful payment scheme.

Otherwise, one ends up with unfounded
assumptions such as, the issuer has to
be an external party!

PS:  from a quick look at the fallacies page
here, http://www.nizkor.org/features/fallacies/
I'd say that the 'bank' fallacy is an Appeal
to Common Practice.  The other one I used
above was an Appeal to Authority ;)

-- 
iang

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