From: hughes@ah.com (Eric Hughes)
Date: Mon, 14 Mar 94 08:29:43 PST
To: cypherpunks@toad.com
Subject: digital cash
In-Reply-To: <199403140422.AA05423@netsys.com>
Message-ID: <9403141619.AA04459@ah.com>
MIME-Version: 1.0
Content-Type: text/plain


>Are you guys going to simply represent dollars with your digital cash
>or will you attempt to create your own currency that may simply be
>converted to/from dollars?

Accounts will be able to be denominated in USA dollars, the central
bank money issued by the USA's own Federal Reserve.  Accounts will
also be able to be denominated in other major currencies traded on the
Foreign Exchange market.  Specifics have not been decided.

We will not be issuing a new currency.

Eric

am
Reply-To: uri@watson.ibm.com
X-Mailer: ELM [version 2.4 PL20]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 3561
Sender: owner-cypherpunks@toad.com
Precedence: bulk

Herb Lin says:
> In the AOL debate between Barlow and Denning, Barlow asserted that Clipper
> increases the gov't capability to do traffic analysis.  Can someone please
> describe the technical basis for this claim?  (No rhetoric please, just the
> technical background...)

Traffic analysis is "who contacts who, when, where from, where to, for
how long and how often".

Today most of the "ordinary public" phone communications are analog/voice.
Thus when a phone call comes in, you have the source of the call (i.e. the
originating phone number), the destination of the call and the voices (you
can analyze them with a reasonable chance to identify the speakers). Today
it's quite feasible  to obscure the identity of the parties  (by using pay
phones, and so)...     But there are no good ways to secure/encrypt analog
voice -  thus no matter what measures you use,  the contents of the dialog
will lay bare.

Another communication mode emerges: digital e-mail and digitized voice.
This may present much harder tracking problem in both party recognition
and location.  Imagine anonymous TCP/IP connection server and sort of a
chain of "remailers"  which  bounce TCP packets  (or should I have said
streams? :-). Plus unbreakable encryption, which deny the eavesdroppers
any chance to pry the contents open... It is possible today.

Now Clipper comes in. Each digital stream coming out of it will have a
tag identifying the source  (in case of dialog each party will present
thus it's chip ID, which uniquely identifies either the party, or it's
location). Note, that when the "voice-remailer" technology picks up (I
assume it will, for the privacy seems to be worth of the price) - even
an "ordinary person" will be able to enjoy the "total" privacy.  While
Clipper can't deny such privacy  to outlaws  (i.e. I can superencipher
the output of Clipper chip, or I can use another encryption altogether
to avoid both decrypting of the contents and identifying with  Clipper
ID) -  it's obvious,  that an "ordinary citizen"  simply won't bother,
just like he/she doesn't go to a train station to make a phone call to
preserve his/her privacy.  No matter how "randomly" will the digitized
[encrypted] voice data stream bounce around through  commercial "voice
remailers", it will have identifying tags attached to it,  allowing to
trace it to it's both end points.

It's not today, that Clipper chip is about - it's the future that it
endangers.  Of course, it all is based on assumptions:

1. Americans want privacy and anonymity (since they also want
   Caller ID, I'm not sure how correct this is).
2. Anonymous "voice remailers" will come up soon  after decent voice
   encryption becomes available cheaply  to the masses,  AND WILL BE
   USED BY GREAT MANY people -  otherwise the chain  "Joe Schmoe has
   called Remailer1, it called Remailer2, ..., it called Jim Schmoe"
   is easily reconstructed  (and you don't even have the benefits of
   random delay before bouncing the pieces off in attempt to confuse
   an eavesdropper whose piece goes out when and where to).
3. The industry will pick up those tools and expand them to the level
   of public phone service (and the gov't won't scare or bribe them
   out from this idea).
4. Traffic analysis is a bad thing and we should deny it to an
   eavesdropper. [Well, is it true? Where's the line betwen "I
   don't really care" and "Now my freedom i in danger"?]
5. People are good (no, I'm joking! :-).
--
Regards,
Uri         uri@watson.ibm.com      scifi!angmar!uri 	N2RIU
-----------
<Disclamer>