[412] in cryptography@c2.net mail archive

home help back first fref pref prev next nref lref last post

Re: Dorothy and the four Horseman

daemon@ATHENA.MIT.EDU (Hal Finney)
Fri Mar 21 15:42:03 1997

Date: Fri, 21 Mar 1997 12:12:05 -0800
From: Hal Finney <hal@rain.org>
To: cryptography@c2.net

Phil Karn <karn@laptop.ka9q.ampr.org> wrote:
>More seriously, shortly after TWA flight 800 went down, I sent email
>to James Kallstrom suggesting that he publish a PGP key in case
>someone wanted to make a confidential tip to the FBI. Since we had
>already met in person, I offered to sign the key. He responded with a
>nice, polite note, declining the offer.

I like the idea of government agencies having their own keys, but by
itself that won't help much with confidential tips.  I doubt that many
tipsters would be deterred by fear that their messages to the FBI would
be read by their ISP's or other eavesdroppers.

What they really need is some form of anonymity.  The FBI should publish
an email address inviting anonymous tips to that address and pledging not
to put pressure on remailer operators to find out where messages came from.
RemOps could then block other law enforcement related addresses and let
things go to this one.

Sure, a lot of the email would be fraudulent or threatening, but probably
anonymous-tip phone lines have the same problem.  The real change needs
to be an institutional one, to stop viewing anonymous messages as a problem
whose solution is to have a way to trace them.  Rather, they should be
seen as an important channel which allows some information to come through
that couldn't be delivered any other way.

My experience as a remailer operator suggests that we are going to have to
rethink how we approach internet anonymity.  I believe the key may be
to deliver messages only to mailboxes which have voluntarily registered
as being willing to receive anonymous information.  This won't eliminate
all the problems, but it would cut out > 90% of the ones we have now.

Hal

home help back first fref pref prev next nref lref last post