At 2:42 PM 7/22/95, Dan Bailey wrote:

On Wed, 19 Jul 1995 11:39:07 -0700 you wrote:

...

At 11:02 AM 7/19/95 PDT, rick hoselton wrote:

...

I want to register the 1-bit key of "1". I expect to send about half my message bits encrypted, the rest will be clear-text.

Oh, go ahead, register 0 also. You'll probably want to switch keys occasionally during sessions.

Actually, why don't we just register our favorite geometric constant, pi? Assuming it's non-repeating, and non-terminating, you're guaranteed that whatever key you end up using will be somewhere in pi.

Reasons this won't work: 1. The Real Reason: It's terminally cute, and terminally cute arguments rarely stand up in court. 2. The Technical Reason: As I recollect, it is unproven that "any sequence of digits will appear in pi someplace." (It may be expected that any finite sequence will eventually appear, but I'm unaware of any proof, and I have reason to suspect such a proof might be impossible.). A wise-ass judge--not that any court in the rational (or irrational) world would ever deal with this--could demand proof. 3. The Legalistic Reason: The "key registration" law would likely be phrased in terms of direct opening of messages, not existential trickery about "the set of all keys." I urge that we deal with key registration on more plausible bases than trickery and sophistry. (How may keys can dance on the head of a PIN, and all.) --Tim May .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@sensemedia.net | anonymous networks, digital pseudonyms, zero 408-728-0152 | knowledge, reputations, information markets, Corralitos, CA | black markets, collapse of governments. Higher Power: 2^756839 | Public Key: PGP and MailSafe available. "National borders are just speed bumps on the information superhighway."