From: Wei Dai <weidai@eskimo.com>
Date: Sun, 7 Apr 1996 12:40:51 +0800
To: Gary Howland <gary@kampai.euronet.nl>
Subject: Re: e$ seigniorage (and is this the cost of untracability?)
In-Reply-To: <199604061906.VAA16731@kampai.euronet.nl>
Message-ID: <Pine.SUN.3.92.960406154934.27655B-100000@eskimo.com>
MIME-Version: 1.0
Content-Type: text/plain


On Sat, 6 Apr 1996, Gary Howland wrote:

>     - What will be the typical time between the withdrawal of ecash
> 	  and it being deposited?

I think this will depend on how easy it is to withdraw ecash.  If the
client software includes an option of automaticly withdrawing ecash from
the bank when you don't have enough ecash to pay for the current purchase
(thereby reducing the time between withdrawal and deposit to zero), then I
suspect most people will use it, even though (anticipating your next
question) this compromises their untraceability.

>     - Does the untraceability of ecash (of all types) rely on a time delay
>       between withdrawal and deposit (I guess it does), and if so, is the
>       interest that the consumer inevitably loses (and the ecash issuer gains)
>       the price the consumer must pay for untracability of transactions?

The untraceability of ecash does depend on a time delay between withdrawal
and deposit.  It's analogous to the fact that the untraceability of
anonymous e-mail depend on a time delay between sending and receiving.  I
think you're probably right that the interest lost is a price the consumer
must pay for untraceability.

One possible way to get around this is to have ecash issuers pay interest
on ecash.  However it requires ecash to be timestamped and therefore
compromises its untraceability.  (Think of the timestamp as a serial
number.)

Wei Dai