NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/sci/crypt/5002

Text File | 1992-11-18 | 3.9 KB | 68 lines

Newsgroups: sci.crypt
Path: sparky!uunet!zaphod.mps.ohio-state.edu!n8emr!colnet!res
From: [e-mail] (Rob Stampfli)
Subject: Re: Demons and Ogres
Message-ID: <[e-mail]>
Organization: Little to None
References: <[e-mail]> <[e-mail]> <[e-mail]>
Date: Thu, 19 Nov 1992 06:03:07 GMT
Lines: 57

In article <[e-mail]> [e-mail] (Terry Ritter) writes:
> Nonsense.  My proposal is that, given the ciphertext, the user
> would be required to make it plain.

This is the gist of what Terry has stated several times.  I am concerned
about such a proposal because it seems to me that there is a dangerous
but unspoken precedent that goes along with it: Aside from the argument
previously espoused that it is too invasive to require one to keep track
of any key they might ever have used (on the supposition that someone in
authority might surreptitiously be recording the transaction and might at
some later point demand the plaintext), this whole line of reasoning appears
to revoke the principle that places the burden of proof on the accuser.
I'm afraid that if we adopt this principle, that using cryptography will
come to be viewed like driving, where certain rights are given up in return
for the privilege to operate a motor vehicle on the public roads.
Unfortunately, I'm afraid that the right we will end up losing as crypto
users will be the presumption of innocence.

Suppose the "annonymous informant", or the "expert" who doesn't know as much
as he thinks he knows, gives the authorities an encrypted message using a key
you have never seen before, but which, he alleges, originates from you.
Suppose the tap of your ISDN line misses a few bytes at the start of your
message.  Under these conditions there is little you can do, even if you
desire to be cooperative.  However, your failure to produce the desired
result would place you in considerable legal jeopardy under what is being
proposed here.

Few people I know believe that the erasure of several minutes worth of
conversation from Nixon's infamous office tapes was accidental, but no one
can prove otherwise, so legally we have to give him the benefit of the doubt.
Imagine how it would look if you had to tell the court "I don't know why, but
my key does not work to decrypt this message."  Few people would be willing
to believe you, either.  Obviously, the contents of the message must be
incriminating, and you must have taken the correct key and destroyed it.
After all, the "experts" surely must have checked and rechecked what they
were doing.

With the complexity of modern technology, I find it extremely dangerous
to place this burden on the defendent, who may well be using something he
knows very little about technically.  On the other hand, if you don't
shoulder him with this responsibility, then you give the unethical
defendant an easy out -- it would be difficult not to accept the "I can't
remember the key" or "I don't know why it doesn't work" defenses, as lame
as they appear, and as unacceptable as they would be to the authorities.
And note that Dr.  Dennings proposal(s) are just as flawed: if someone
is going to lie, it hardly behooves them to register all their keys up front.

So, where does this leave us?  The obvious extremes are to ban all strong
cryptography outright (and make it a crime in itself to use it), or to
allow it to be used as a privacy tool of a free society, while accepting
the consequences that it may be impossible to convict someone who is clever
in the manner by which they refuse to surrender their key.  Obviously, I am
in the latter camp, but there are strong entities in this country that
would undoubtedly weigh the evidence and favor the other approach.
--
Rob Stampfli  [e-mail]      The neat thing about standards:
614-864-9377  HAM RADIO: [e-mail]  There are so many to choose from.