From: smb@research.att.com
Date: id AA03087sendmail 5.67/QC-subsidiary-2.1 via SMTPSun, 18 Apr 93 13:14:19 -0700 for karn
To: gong@csl.sri.com, karn
Subject: correction from Hellman
Message-ID: <9304182014.AA01706@qualcomm.com>
MIME-Version: 1.0
Content-Type: text/plain


Here's a follow-up note from Hellman, in response to a question I sent
(and also in response to my request to post his original note to
netnews).

------- Forwarded Message

Date: Sun, 18 Apr 93 11:41:42 PDT
From: "Martin Hellman" <hellman@isl.stanford.edu>
To: smb@research.att.com
Subject: Re: Clipper Chip

It is fine to post my previous message to sci.crypt 
if you also post this message with it in which:

1. I ask recipients to be sparse in their requesting further info 
from me or asking for comments on specific questions. By
this posting I apologize for any messages I am unable to
respond to. (I already spend too much time answering too much
e-mail and am particularly overloaded this week with other
responsibilities.) 

2. I note a probably correction sent to me by Dorothy Denning.
She met with the person from NSA that
I talked with by phone, so her understanding is likely to
better than mine on this point: Where I said the transmitted
info is  E{ E[M; K], E[K; UK], serial number;  SK}
she says the message is not double encrypted. The system
key (or family key as she was told it is called) only encrypts
the serial number or the serial number and the encrypted
unit key. This is not a major difference, but I thought it
should be mentioned and thank her for bringing it to
my attention. It makes more sense since it cuts down
on encryption computation overhead.

------- End of Forwarded Message