Eric Hughes points out that the blind signatures used to provide anonymity in Chaum's digital cash schemes are patented (by Chaum himself). This is a problem for an official, legal, profit-making business which wants to engage in electronic banking. However, such a business would face many other problems. Chaum's digital cash system could be construed as using the RSA algorithms, since it is in effect an RSA signature which makes the cash unforgeable. So a license for RSA would have to be obtained as well. (RSA licensing would also be needed for secure communication between the bank and its customers.) In addition, the acceptable use policies of NSFNET, which would probably have to be involved in most communications with the bank, are inconsistent with this kind of commercial activity, from what I understand. I believe new policies are in the works to allow commercial activities on the net, but these again will involve licensing costs. There is also the question of the legality of private cash in any form, electronic or otherwise. Nobody seems to have hard evidence on this. On the one hand, people can legally exchange certain types of securities, and they could perform services for each other in return for such exchanges. This is legal, but they are supposed to report the income on their tax returns. Our digital cash would seem to fit this model. (But are such securities transfers as untraceable as digital cash? Perhaps not.) On the other hand, I read several years ago that casino chips in Las Vegas were starting to be used as cash substitutes, but that the government cracked down on this practice. Perhaps this was too conducive to money laundering, especially with the reputed underworld activity in that city. I do think, though, that an informal digital cash system, presented as a research project or an educational game, would be able to slip between the cracks of the legal system, much as PGP has done. And I think this presentation would be legitimate. Digital cash is new (actually, nonexistant, as far as I know), and any use of it would by its very nature be research and be educational. I would suggest that anyone who proposes to implement such a game might want to consider releasing it anonymously (actually, pseudonymously). Sign the release with a PGP or RIPEM key, and let that be your pseudonym. Let people post messages to alt.security.pgp or some other newsgroup to discuss it, so that you can read them without revealing yourself, as proposed by Yanek. Post your replies anonymously using our remailers. This way there won't be a single-point target for anyone wishing to punish users of the game. Hal 74076.1041@compuserve.com