🚀 go-pugleaf

Terry Ritter raises the question of the forced discovery of keys AFTER the
legal system has become engaged. I believe the laws related to discovery
are quite broad. In a civil suit, for example, either party has the right
to subpoena any papers of the other they think reasonably relevant. The
only protection is a suppression order by the judge, on grounds of trade
secrets, etc. Such an order does not prevent either party from seeing
the papers, but simply seals them so they cannot be used nor disclosed
to third parties, except for purposes of the lawsuit.

It would seem to me that one could force the disclosure of keys quite
easily, given that background. Anyone could, for example, sue someone
for damages based on some cause, and then use discovery to get
whatever they want. Of course it's after-the-fact, but it makes
public key systems particularly vulnerable, since once they have your
private key (if you don't change it often, or if you done have classes
of private keys for different activities), they can read everything.
Since the whole value of a public key system is to have a single key
set so people can send encrypted traffic to you, anyone "up to no
good" would probably more likely use a private key system, with many
different keys.

Perhaps a smart judge would permit you to furnish plaintext instead of the
key, though likely not, since you couldn't "prove" that was the real plaintext
without revealing the private key. Maybe there's a way to authenticate the
plaintext via signatures, without having to reveal your private key--dunno.

I've always thought the protections in the discovery rules to be very weak.
Once an adversary party knows confidential (sealed) information, unless
it's something clean-cut, such as a unique invention that cannot be duplicated
except via your method, the cat is out of the bag and it's very hard to prove
disclosure or improper use of sealed information.

-- 
David Sternlight
(pgp 2.0 and ripem public keys available on request)