From: "Perry E. Metzger" <perry@piermont.com>
Date: Tue, 28 May 1996 18:39:46 +0800
To: cypherpunks@toad.com
Subject: Re: net-based key archival
In-Reply-To: <9605280603.AA24810@anon.penet.fi>
Message-ID: <199605280732.DAA04752@jekyll.piermont.com>
MIME-Version: 1.0
Content-Type: text/plain



an116512@anon.penet.fi writes:
> I just brought these files down + a few others in the same directory that
> he didnt tell us about (you can see them with netscape by leaving off the
> last part. Check out MKCS.PS and others. They take a while to see because
> you need a postscript viewer but if you can read them they are
> enlightening abot this guys character. It looks like AT&T is in the
> key escrow biz, folks!!! Maybe someone should scan them in and post
> them to the list, so we know what the enemy is doing.
> 
> The MKCS.PS file is about how to make a code for other people to use so
> that you can break. And then they say dont worry the government would never
> do this!!! Riiiiggggght.

God, your are an imbecile. You couldn't even be bothered to read the
document you are yelling about, could you.

The MKCS abstract explains the result that Blaze et al arrived at
which shows that generalized systems with back doors in them are
roughly equivalent to public key systems -- that is, any generalized
Master Key Cryptosystem can be used as a public key system. Thus,
although it is possible that there are faster techniques available, it
would appear that to to design a cipher with a back door, we would
have to use techniques that are currently thought to be slow. A lot of
the point was that ciphers like DES are unlikely to have "master key"
style back doors. This result is about the opposite of what you
trumpet.

They come up with some other interesting results about master keyed
systems. None of them are "AT&T is in the key escrow biz" revelations.

Might I suggest that in the future you have someone jackhammer your
head out of the concrete block it is encased in before you make
pronouncements?

Perry