DalSemi: Add-Only Memory for Storage of Digital Cash

A press release breezed through the list a while back about some
"decoder rings" -- basically, a nonvolatile RAM chip embedded in a
Jostens ring.

On further inspection, it looks to be a superior product to the
DataKey products I remember seeing about ten years ago, targeted at
this same sort of niche. The part in question is much smaller -- it
fits in a ring, after all, or on the usual sorts of key fobs and
employee badges -- and establishing electrical contact to transceive
data is trivial.

Cypherpunks relevance? Twofold. One, the Touch MultiKey
(ftp://ftp.dalsemi.com/pub/datasheets/1991.ps), which promises to hold
three 384 bit blocks under 64-bit passwords. The device will transmit
the stored data under the correct password, and "random bits" under
all other passwords. No crypto here, though, just a simple on-chip
comparison with a stored password. So I assume the determined opponent
with physical access can extract the info; but better that than store
your private key on a publically accessible machine, no?

Secondly, an app note: "Use of Add-Only Memory for Secure Storage of
Monetary Equivalent Data" (ftp://ftp.dalsemi.com/pub/datasheets/app84.ps).
A creative idea based on the peculiar nature of the EPROMs* that are
also available in this form factor. In particular, one bits may be
burned to zeros, but not vice versa; so why not burn particular bits
to indicate credits and debits?

Knowing that this scheme depends on keeping an attacker from guessing
which bits to burn, they use the unique serial number to uniquely
permute the bits, so that an attacker is as likely to burn a "debit"
bit as a "credit", and far likelier to burn out-of-sequence so that
the monkey business is apparent to any vendor.

And there's another item of note: each chip has a unique, etched,
machine-readable serial number. What are the bets that Dallas
Semiconductor can tell you who purchased that chip? Well, so much for
an anonymous payment scheme based on *this* product.

Still, if I had to choose a place to keep a secret key, I'd choose my
knuckle over my key ring, let alone a floppy disk, PDA, or portable
computer.

nathan
* Gee, back in my day EPROMs were Eraseable; these folks mean instead
  to indicate an Electrically Programmable chip, which sounds like a
  good old PROM to me. Ahhh, acronyms...

-----BEGIN PGP SIGNED MESSAGE-----

In list.cypherpunks, loofbour@cis.ohio-state.edu writes:

> And there's another item of note: each chip has a unique, etched,
> machine-readable serial number. What are the bets that Dallas
> Semiconductor can tell you who purchased that chip? Well, so much for
> an anonymous payment scheme based on *this* product.

Dallas might be able to tell who purchased them, but it's likely to have
been someone like Hamilton-Hallmark (a major parts distributor).
Whether the distributors are going to maintain the audit trail is
questionable at best.  It only takes one distributor to break the
chain of traceability, and audit trails cost money.

> * Gee, back in my day EPROMs were Eraseable; these folks mean instead
>   to indicate an Electrically Programmable chip, which sounds like a
>   good old PROM to me. Ahhh, acronyms...

This comes from the fact that PROMs are typically mask-programmed at the
fabrication stage.  EPROMS are programmed in the field.  And yes, if you
could get the top of the chip off cleanly, you could erase and reprogram
them.  (not likely, and hardly undetectable, but it's ever so slightly
possible)  More often, I see these devices called OTP (for One Time
Programmable).  In the more standard types (27C256, for example),
they're a fraction of the cost of the erasable ones.  Ceramic cases and
quartz windows are spendy.
- -- 
           Roy M. Silvernail     [ ]      roy@cybrspc.mn.org
PGP Public Key fingerprint =  31 86 EC B9 DB 76 A7 54  13 0B 6A 6B CC 09 18 B6
                Key available from pubkey@cybrspc.mn.org

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBMIQv0Rvikii9febJAQEeKAP+LbgyZ/60fGuVICZEqM+Rv34GhEA6a/vg
cKPbKCazUVg8bEBod3mqbHfssjDgD47PcAai8uM3ALmki/TI3DfI6FLbZr7aCpa8
PSNFDTEpmRDpnm5xpbZa/5O1aLdXLX6ps8OGsg0YjY1hvFQCn5tymW9GjhOXrkXS
s698T5nEoQI=
=LoNA
-----END PGP SIGNATURE-----

Nathan Loofbourrow writes:
>And there's another item of note: each chip has a unique, etched,
>machine-readable serial number. What are the bets that Dallas
>Semiconductor can tell you who purchased that chip? Well, so much for
>an anonymous payment scheme based on *this* product.

Some enterprising cypherpunks can buy a bunch and resell 'em for cash.

On Tue, 17 Oct 1995, Scott Brickner wrote:

> Nathan Loofbourrow writes:
> >And there's another item of note: each chip has a unique, etched,
> >machine-readable serial number. What are the bets that Dallas
> >Semiconductor can tell you who purchased that chip? Well, so much for
> >an anonymous payment scheme based on *this* product.
> 
> Some enterprising cypherpunks can buy a bunch and resell 'em for cash.
> 
And then sell the serial address for cash.
How do I know that they (this means YOU! :> ) don't work for the company?
(which is a contracter for, blackmail-victim of, tentacle of NSA, CSE, 
TCMAY, Purplenet, your fairy-stepmother ...)

Of course there's always the mail-drop & forward-net... (anyone one know
maildrop company addresses, the possible (il)legality of these things or 
any reported monitoring cases? A physical remailernet might come to matter in an
economy where the "split" between an untaxable info economy and a 
taxable physical economy become pronounced. Though it's kinda hard to encrypt
and reorder packages.)

Someone said:
> > Some enterprising cypherpunks can buy a bunch and resell 'em for cash.
> > 
> And then sell the serial address for cash.

CypherChip Party. Pick up one chip from pile and leave one chip, or
leave $$.

-- 
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

s1018954@aix2.uottawa.ca writes:
>On Tue, 17 Oct 1995, Scott Brickner wrote:
>
>> Nathan Loofbourrow writes:
>> >And there's another item of note: each chip has a unique, etched,
>> >machine-readable serial number. What are the bets that Dallas
>> >Semiconductor can tell you who purchased that chip? Well, so much for
>> >an anonymous payment scheme based on *this* product.
>> 
>> Some enterprising cypherpunks can buy a bunch and resell 'em for cash.
>> 
>And then sell the serial address for cash.
>How do I know that they (this means YOU! :> ) don't work for the company?
>(which is a contracter for, blackmail-victim of, tentacle of NSA, CSE, 
>TCMAY, Purplenet, your fairy-stepmother ...)

Big deal.  NSA now knows "someone bought address X".

That's the great thing about cash you know... anonymity.