[148869] in cryptography@c2.net mail archive
Re: [Cryptography] What is a secure conversation? (Was: online
daemon@ATHENA.MIT.EDU (John Gilmore)
Wed Jan 1 15:42:59 2014
X-Original-To: cryptography@metzdowd.com
To: Phillip Hallam-Baker <hallam@gmail.com>
In-reply-to: <A9633EF0-1917-45D3-910E-1E413A31F149@gmail.com>
Date: Wed, 01 Jan 2014 11:25:36 -0800
From: John Gilmore <gnu@toad.com>
Cc: Cryptography <cryptography@metzdowd.com>, Arnold Reinhold <agr@me.com>
Errors-To: cryptography-bounces+crypto.discuss=bloom-picayune.mit.edu@metzdowd.com
>> The U.S. Government's first priority should be to do this at our nuclear weapons labs, before our bomb design codes (in Fortran, no doubt) wind up on BitTorrent.
>
> The nuclear codes leaked long ago. I know the navy one from the 60s
You are talking about two different things.
The "bomb design codes" are software programs built to simulate
nuclear reactions in particular physical constructions. These have
the original Manhattan Project calculations (done on punched cards,
with tabulators, as described in Feynmann's autobiography) as remote
ancestors.
The "nuclear codes" ("launch codes") are a short sequence that arms the
missiles before sending them into the sky. The "key" of a security
system that keeps particular bombs from exploding.
Both of them need a lot of other infrastructure to be useful to an
adversary. Launch codes from the '60s are probably declassified now
anyway. Nuclear simulation software in FORTRAN from the '60s would
still be useful in designing nuclear bombs. (Actually building them,
getting them to targets, and detonating them would be big challenges.)
John
_______________________________________________
The cryptography mailing list
cryptography@metzdowd.com
http://www.metzdowd.com/mailman/listinfo/cryptography
