🚀 go-pugleaf

In article <1992Nov3.090942.1626@guvax.acc.georgetown.edu> denning@guvax.acc.georgetown.edu writes:

	[ a list of benefits of wiretapping for law enforcement ]

>According to the FBI, the entire hierarchy of [Organized Crime] has been
>neutralized or destabilized through the use of electronic surveillance and
>30 odd years of success would be reversed if they lost the ability to
>conduct court-authorized electronic surveillance.

	[ with similar statements re: drug trafficing ]

These sound like arguments against making knowledge of strong encryption
public and against making any encryption devices or software publicly
available.

>I believe I said this before, but let me remind everyone that the
>government has not proposed that keys be registered.  I believe that Ron
>Rivest was the first to make this suggestion.  I made it at the NCSC and
>Silvio Micali has come up with a scheme that is even more foolproof.  I do
>not know if any government agency will pick up on our suggestion or if
>they'll suggest another.

My question remains: how will key registration preserve the ability to
wiretap?  If the damage done by wiretaps has been so extensive, unless the
OC and drug traffic figures are incredibly stupid, they will have acquired
or will be about to acquire off-the-shelf phone scramblers and message
encryption devices.  Granted, there may be some who are of the old school
-- refuse to learn about PCs and e-mail -- but that gang will die off
(partly by wiretapping, it seems).

If I were the gov't with all these facts/figures in mind, I'd be arguing
against publication of cryptologic results (ala the attempt in 1978) and
would consider a key registration proposal too feeble to be worth anything
-- except, as Welchman and others point out from WW-II, if a secret runs
around a secure net it will also leak out to less secure nets and might be
picked up (or deduced) from there.

Therefore, if you can tap at will, you might be able to piece together the
information which good encryption has denied you.

However, your claim is that the gov't doesn't want (and can't afford) to do
this -- so I don't see how key registration can help.

>Are there other approaches that would give you greater assurance against
>the government and still allow for meeting the legitimate needs of law
>enforcement?  If so, I'd love to hear them.

I, too, would love to see sci.crypt turn its collective head to this topic.
In particular, there are protections we would like against abuses by the
government -- COINTELPRO, Watergate and Iran Contra to name a few.

Cryptography gives us a means for controlling access to information.

We want to preserve privacy of citizens.  We want to prevent secret
conspiracies within the White House, for example.  We want to allow the FBI
to get evidence against mob bosses and true drug traffickers but not
against the two of us (without our knowledge and chance for legal or
political remedy).  We want to prevent the kinds of abuses which have
occurred in the past and which have led us ever closer to an imperial
presidency and the threat of dictatorship in America.


If we were to assume, for example, that all comm lines and all data were
encrypted, could we design a protocol for handling the access to that
data such that all rightful privacy were preserved and all illegal activity
were exposed?  We would have to exclude up front any assumption of the
form that the gov't (or anyone else) is honest.

This sure sounds like the domain of cryptographic protocols, to me.

-- 
-- <<Disclaimer: All opinions expressed are my own, of course.>>
-- Carl Ellison						cme@sw.stratus.com
-- Stratus Computer Inc.	M3-2-BKW		TEL: (508)460-2783
-- 55 Fairbanks Boulevard ; Marlborough MA 01752-1298	FAX: (508)624-7488