Xref: utzoo comp.admin.policy:279 comp.unix.admin:2125
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!wri!wri.com!ben
From: ben@wri.com (Ben Cox)
Newsgroups: comp.admin.policy,comp.unix.admin
Subject: Re: E-mail Privacy
Message-ID: <BEN.91Jun5112150@dragonfly.wri.com>
Date: 5 Jun 91 17:21:50 GMT
References: <51171@prls.UUCP> <1991Jun3.175631.1451@sci34hub.sci.com>
	<1991Jun4.144731.685@forwiss.uni-passau.de>
	<1991Jun4.194406.1366@qualcomm.com>
Sender: ben@wri.com (Ben Cox)
Organization: Wolfram Research, Inc.
Lines: 26
In-Reply-To: rdippold@cancun.qualcomm.com's message of 4 Jun 91 19: 44:06 GMT

rdippold@cancun.qualcomm.com (Ron Dippold) writes:

> Crypt makes use of the Data Encryption Standard (DES), an encryption
> technology that is supposedly unbreakable without spending nearly
> infinite amounts of computer time (although many believe that the
> National Security Agency purposely weakend the specifications to the
> point where they _can_ decode it).

As has been pointed out by others, crypt(1) does not use the DES
encryption standard.  There is a DES implementation out, though, that
someone else mentioned in a post.

Something interesting to note, though, appears on page 450 of "The
Unix System Administration Handbook" by Evi Nemeth, Scott Seebass, and
Garth Snyder (Prentice Hall, 1989, reprinted without permission) in a
footnote:

   Evi broke the Diffie-Hellman key exchange often used with the DES
   encryption method using a HEP supercomputer in 1984.  Although the
   DES algorithm is quite complicated, nothing crypted with DES can be
   considered 100% secure.  The U.S. government has been (rightly?)
   accused of blocking adoption of encryption standards that cannot be
   broken by the NSA.

-- Ben Cox
   ben@wri.com