Re: Actively Sabotaging Clipper and Capstone?

Tim May posted to cypherpunks:


> Hey, I've just been told in e-mail that my $200,000 figure for the
> Clipper keys (a tape or compilation of the ones that are held in
> escrow) is way too low, probably by two orders of magnitude.
>
> Maybe so, as having these keys could mean a lot.
>
> But my point is that nearly any such figure will represent an
> incredible temptation. Such is the risk of any centralized system in
> which a master key (or set of escrowed keys) unlocks such valuable
> information.


This is exactly the same argument that corrupts the PEM certification
scheme. While hierarachical chains of command are reflexivly the first
idea in any military or bureaucratic employee, the existance of a
super-valuable "master certification certificate" that is valuable will
directly make it extremely valuable. Anything of sufficient value will
be compromised by someone willing to pay a sufficient value, break a
kneecap, etc.

Once a valuable, high level certificate is broken, then all decendant
certificates are broken. The "web of trust" is the only workable solution.

Pat

Pat Farrell      Grad Student                 pfarrell@gmu.edu
Department of Computer Science    George Mason University, Fairfax, VA
Public key availble via finger          #include <standard.disclaimer>

Hm... it seems to me, that if the government is going to have these keys, 
it won't be too long before they are all posted to the net... considering 
how good the .gov and .mil is at keeping secrets, we shouldn't woory :)

___________________________________________________________________________
|---===================================--|     /\     |  |  \ |_ _\ \  / |
|---Colin Titus Orion Xavier Chandler----|    \\ \    |  | .  |  | >  <  |
|---===================================--|   \ \\ /  \__/ _|\_|___|_/\_\ |
| _____                                  |  / \/ / /		         |
|/\  __ \         __  "What year is it?" | / /   \//\   "If it's not a   | 
|\ \ \/\ \  _ __ /\_\    ___     ___     | \//\   / /	Sun, it's not a  |
| \ \ \ \ \/\`'__\/\ \  / __`\ /' _ `\   |  / / /\ /  	   computer."    |
|  \ \ \_\ \ \ \/ \ \ \/\ \L\ \/\ \/\ \  |   / \\ \  .__          __     |
|   \ \_____\ \_\  \ \_\ \____/\ \_\ \_\ |    \ \\   |_. | | |\ |  -|    |
|    \/_____/\/_/   \/_/\/___/  \/_/\/_/ |     \/    __| I_| | \| __|/160|
+________________________________________+_______________________________+
|  Colin Chandler |"It can only be accountable to *human* error."-HAL9000|
|  (415) 388-8055 | orion@crl.com, wizard @ BayMOO (mud.crl.com 8888)    |
|________________________________________________________________________|

On Mon, 14 Feb 1994, Pat Farrell wrote:

> Tim May posted to cypherpunks:
> 
> 
> > Hey, I've just been told in e-mail that my $200,000 figure for the
> > Clipper keys (a tape or compilation of the ones that are held in
> > escrow) is way too low, probably by two orders of magnitude.
> >
> > Maybe so, as having these keys could mean a lot.
> >
> > But my point is that nearly any such figure will represent an
> > incredible temptation. Such is the risk of any centralized system in
> > which a master key (or set of escrowed keys) unlocks such valuable
> > information.
> 
> 
> This is exactly the same argument that corrupts the PEM certification
> scheme. While hierarachical chains of command are reflexivly the first
> idea in any military or bureaucratic employee, the existance of a
> super-valuable "master certification certificate" that is valuable will
> directly make it extremely valuable. Anything of sufficient value will
> be compromised by someone willing to pay a sufficient value, break a
> kneecap, etc.
> 
> Once a valuable, high level certificate is broken, then all decendant
> certificates are broken. The "web of trust" is the only workable solution.
> 
> Pat
> 
> Pat Farrell      Grad Student                 pfarrell@gmu.edu
> Department of Computer Science    George Mason University, Fairfax, VA
> Public key availble via finger          #include <standard.disclaimer>
>

You may have received a few copies of the post in the In-reply-to:
field above.  I've removed the apparent cause of the mail loop.  Not
to worry.

Eric