🚀 go-pugleaf

I wonder how this proposal would apply to other kinds of encryption than
public-key.  For secret-key encryption, I suppose the requirement would
be to reveal the secret key to the trusted agency.  Some people might
have a list of keys which they will change frequently.  Again, the list
would have to be registered, presumably.

This brings us to the one-time pad.  I suppose the entire pad would have to
be registered with the government in order to be used.  Random-number
generators might become classified as munitions since they could be used
to support unbreakable cryptography (if the communicants could exchange
random number tables, such as on a CD-ROM).  Possession of unregistered
random number tables could be incriminating evidence.

There have been some proposals for anonymous email, in which the source
and destination of message traffic would be hidden.  David Chaum has a
couple of papers on this topic.  One of his proposals, the "DC-net", can be
implemented using random number tables similar to one-time pads to provide
unbreakable message source confidentiality.  Again, the government would
presumably have to require the pads to be registered so that messages could
be traced back, if necessary, to the criminals who are sending them.  Or
maybe it would be simpler to just outlaw this whole technology.

I wonder about other areas of cryptographic research: new encryption
algorithms; zero-knowledge proof systems; key exchange systems;
cryptographically secure random number generators; signature systems; etc.
Will it become routine to analyze these, not in terms of their strengths
and weaknesses, but in terms of how convenient they will be for governments
to control?  Diffie-Hellmann key exchange, for example, where one generates
new encryption keys dynamically at the time of use, would appear to be
outside the boundaries of these proposed registration systems.  Do we then
have no choice but to forbid its use?

Hal Finney