From: Paul Robichaux <paul@ljl.COM>
Date: Wed, 5 Jun 1996 12:14:59 +0800
To: cypherpunks@toad.com
Subject: Re: Markoff in NYT on NTT/RSA chip
In-Reply-To: <Pine.3.89.9606032128.A10202-0100000@well>
Message-ID: <v03007101adda1bc74f88@[206.151.234.118]>
MIME-Version: 1.0
Content-Type: text/plain


There are still a few unclear (to me) points about this story:

1. CWD says that RSA's Japanese subsidiary developed the chips. Markoff
says NTT did. It sounds to me like NTT is the more likely developer, since
they have a great deal of silicon design experience.

2. Markoff's story says that NTT doesn't have licenses to use RSA
technology (the quote from Junichi Kishigami), but the bit about safely
swapping private-key info makes me wonder if instead they're using DH-- the
patent for which expires next year. A telecom chipset could make productive
use of DH for key exchange at call setup.

3. Once you import an ITAR item, its export becomes controlled. Importing
the NTT chipset for use, say, in a Motorola cell switch (made in IL) would
seem to be problematic. Buying chips in Japan for shipment to Moto's phone
factories in Singapore and Malaysia, however, would appear to be OK. Under
the present export regs, it's not clear to me how many US manufacturers
would adopt this chipset since it doesn't give them any export relief. If
the NRC's recommendations are adopted, that still won't help, since
triple-DES is over their recommended limit.

The _good_ news is that US manufacturers could put these chipsets into
domestic-only products, and I sincerely hope they will.

-Paul


--
Paul Robichaux                    LJL Enterprises, Inc.
paul@ljl.com                      Be a cryptography user. Ask me how.