Re: Voice/Fax Checks

At 10:52 PM 7/20/94 -0400, Duncan Frissell wrote:

>"Don't bother.  Take out the check you were going to send me, read me the
>routing code and check number on the bottom.  Give me your name and address
>and the bank's name and address as they appear on the check, the amount you
>will pay and the date.  I'll collect that check electronically without you
>having to bother to send it."

This is exactly the problem we're having with identifying a market for
digital cash. There's no unique selling proposition besides privacy. There
are too many real good substitutes, like this one for checks. E-mail with
the above information in it can be encrypted and signed, and would be
secure enough to make a real good check in its own right. This is like my
favorite quote (in InforWorld) about Macs: "It seems that 85% of the market
will settle for 75% of a Macintosh."

By no means take this to mean that digital cash isn't going to make it.  I
figure all e$ now, including the encrypted check above, is kind of like
aviation was in the beginning.  It's really cool that that it works, we can
make some pretty good guesses as to its possible uses, but nobody's built
the "DC-3" which proves once and for all its commercial necessity.  I
expect that the only way to find out whether digital cash is gonna make it
on it's own is when someone risks a small pile and implements it.  Let the
devil take the hindmost, more guts than brains, and all that.  It looks
like maybe that's what Chaum and Co. is going to try to do, with this test
of theirs.  Has anyone out there been contacted about it yet?

Cheers,
Bob


-----------------
Robert Hettinga  (rah@shipwright.com) "There is no difference between someone
Shipwright Development Corporation     who eats too little and sees Heaven and
44 Farquhar Street                       someone who drinks too much and sees
Boston, MA 02331 USA                       snakes." -- Bertrand Russell
(617) 323-7923

> At 10:52 PM 7/20/94 -0400, Duncan Frissell wrote:
> 
> >"Don't bother.  Take out the check you were going to send me, read me the
> >routing code and check number on the bottom.  Give me your name and address
> >and the bank's name and address as they appear on the check, the amount you
> >will pay and the date.  I'll collect that check electronically without you
> >having to bother to send it."
> 
> This is exactly the problem we're having with identifying a market for
> digital cash. There's no unique selling proposition besides privacy. There
> are too many real good substitutes, like this one for checks. E-mail with
> the above information in it can be encrypted and signed, and would be
> secure enough to make a real good check in its own right. This is like my
> favorite quote (in InforWorld) about Macs: "It seems that 85% of the market
> will settle for 75% of a Macintosh."

The selling point for digital cash is that it has a low transaction cost
and can easily be used for extremelly small transactions. If agent A and
agent B want to do business without bothering their owners, you had better
have some robust digicash.

Fellow Cypherpunks,

(Sorry to break in on the flames about bimbos, tentacles, and quantum
computers, etc. And since I have nothing to say about new releases of
PGP 2.6ui, CFS, WinPGP, or PGS, I'll focus on some things that
interest me these days.)

solman@MIT.EDU wrote:

> The selling point for digital cash is that it has a low transaction cost
> and can easily be used for extremelly small transactions. If agent A and
> agent B want to do business without bothering their owners, you had better
> have some robust digicash.

Very fine granularity digital cash--sub-cent levels, even
sub-millicent levels--could have many uses. Multiple transactions,
transations by "agents" (like Telescript will reportedly have), etc.

Cypherpunks should be aware of several tie-ins that some of our
members are working on:

* Norm Hardy and Dean Tribble have been working on a scheme called
"Digital Silk Road," or DSR, in which fractional-cent payments may be
made without incurring the full overhead of a commlink to a bank
clearinghouse, for example. (As communication charges drop, the
overhead cost of a clearinghouse call could be small enough not to
matter, but not for a while....and I'd still worry about the speed of
light delays if nothing else!).

- a version of their DSR work should be available in the usual places
(Netcom's ftp site, the ftp.csua.berkeley.edu site, and various
Cypherpunks-oriented URLs that get posted here often).

* Mark Miller, Eric Drexler, and others have worked on a scheme they
call "agorics," for computer-mediated markets, auctioning of computer
resources, etc.  This developed from work with Xanadu and AMIX, and
other places.  (Ironically, my last major project at Intel, in 1986,
was the explication of a 'Frame-Based Manufacturing System,' in which
scarce wafer fab resources are bought and sold in a manufacturing
ecology.  Miller and Drexler visited my old group a year or so after I
left to talk to them....by this time I also knew Miller and Drexler in
other contexts.)

- Mark will be speaking on the Agorics Project, and the connections to
crypto, at the next Cypherpunks meeting.

(Miller, Tribble, Hardy, and others are working on several projects of
potential interest to us: the "Joule" programming language (built in
Smalltalk, as I recall, but eventually to be ported to a faster and
lower footprint form), the "CORBA-mite" (I hope I got the spelling
right...it's a pun) extension to C++, and some network allocation work
involving special kinds of auctions.

(The common thread is one of market processes, such as the George
Mason U. folks are interested in, the economic theories of F. Hayek
that underly modern libertarian economics, and the very common
sensical notion that things have costs and that agent who want things
more than other agents should expect to pay more. "Computational
ecologies" is another buzzword, and there are obvious resonances with
"ariticial life." In fact, it was at the first A-LIFE conference, in
1987, that I met Mark Miller--I already knew Drexler.)

* Software payment schemes, including "superdistribution" and the
various ideas of Brad Cox, Peter Sprague, etc., are very much related
to fine granularity digital cash.

* The amazing new book by Kevin Kelly, "Out of Control," has a chapter
devoted to digital money. Pick it up at your local bookstore--it
should be in even the tiniest of stores--and at least skim the chapter
on digital money. Don't be scared off by the opening line of the
chapter, in which yours truly compares strong crypto to a
shoulder-fired Stinger missile! (For the curious, Kevin used his
"Whole Earth Review" article from last summer as the basis for this
chapter.)

* In a related note, we discussed this book at the most recent
"Assembler Multitudes" gathering in Palo Alto. This group meets to
discuss the implications of technology, with a historical focus on
nanotechnology. Ted Kaehler, one of the creators of Smalltalk at Xerox
PARC in the 70s, and now working with Alan Kay at Apple, says this
book is the most exciting thing he's seen in many years. I mostly
agree.

* I also described my ideas on a "protocol ecology," a soup of agents
(named after our crypto friends Alice, Bob, Charles, Eve, and so on)
interacting with cryptographic primitives and combining methods and
behaviors. (Basically, Koza-style genetic programming, but done with
method combination on primitives, rather than LISP- or C++-style
mutation and rewriting of code.)

I suspect this short description is not enough to make clear what I
have in mind...it took me an hour to flesh out the explanation to Ted
(and to others present, including Nick Szabo). It may have relevance
to digital cash schemes, and attacks and defenses, in terms of
evolving complex interactive protocols. (Think of Doug Lenat's
Eurisko, from the early 1980s.) I'll write more on this, and the work
I've been doing with SmalltalkAgents, when it's further along.

* Finally, some of our attendees at the local Cypherpunks
meetings--I'm thinking specifically of Scott Collins and Fen
LeBalme--have experience at General Magic and Apple with "Telescript"
and agents. Little word is leaking out on Telescript--our own Peter
Wayner could say little concrete about it in his article for "Byte"
several months back. But it could be very important.

So, there's a lot of exciting stuff going on. I'm convinced that the
vaunted tongue-twister of the 1960s and 70s, "mutually suspicious
cooperating agents," will come to the fore again. (If you don't get
this reference, sorry.)

Reputations, agents, agorics, and digital money. Living in perfect
harmony. I hope.

(I now return control of the Cypherpunks Channel to its normal
programming schedule of insults, babes, political correctness
lectures, rants about, to, and from Detweiler, and, on tonight's viewing
schedule, "Cayman Islands H.E.A.T."(*).

--Tim May

(* Who else considers it not a coincidence that the babelicious Alison
Armitage shares a last name--or close--with a denizen of Bill Gibson's
world? The cypher/cyberpunk connection we've all been hunting for?
Cyphermancer?)


-- 
..........................................................................
Timothy C. May         | Crypto Anarchy: encryption, digital money,  
tcmay@netcom.com       | anonymous networks, digital pseudonyms, zero
408-688-5409           | knowledge, reputations, information markets, 
W.A.S.T.E.: Aptos, CA  | black markets, collapse of governments.
Higher Power: 2^859433 | Public Key: PGP and MailSafe available.
"National borders are just speed bumps on the information superhighway."

Robert Hettinga writes
> ... the problem we're having with identifying a market for
> digital cash. There's no unique selling proposition besides privacy. There
> are too many real good substitutes, like this one for checks. E-mail with
> the above information in it can be encrypted and signed, and would be
> secure enough to make a real good check in its own right.

All existing substitutes are either insecure (credit cards) or involve
excessive labor and transaction costs.

Electronic transactions will take off like a rocket once they 
*undercut* existing methods.

As yet, our mail encryption interface is still bad.  Convenient
crypto cash must come after convenient crypto mail.


-- 
 ---------------------------------------------------------------------
We have the right to defend ourselves and our
property, because of the kind of animals that we              James A. Donald
are.  True law derives from this right, not from
the arbitrary power of the omnipotent state.                jamesd@netcom.com