1997-10-06 - PGP5 uses El Gamal not DH

Header Data

From: Adam Back <aba@dcs.ex.ac.uk>
To: aldius@mindless.com
Message Hash: 81651d21a756883256fb00d070652443054e1ff9b008bc15355ff36fe7c60e13
Message ID: <199710060817.JAA00800@server.test.net>
Reply To: <199710060054.RAA02677@toad.com>
UTC Datetime: 1997-10-06 11:03:08 UTC
Raw Date: Mon, 6 Oct 1997 19:03:08 +0800

Raw message

From: Adam Back <aba@dcs.ex.ac.uk>
Date: Mon, 6 Oct 1997 19:03:08 +0800
To: aldius@mindless.com
Subject: PGP5 uses El Gamal _not_ DH
In-Reply-To: <199710060054.RAA02677@toad.com>
Message-ID: <199710060817.JAA00800@server.test.net>
MIME-Version: 1.0
Content-Type: text/plain




Aldius <aldius@mindless.com> writes:
> The reason for having two key types for PGP5 is that DH is only capable
> of key distribution, not signing.  

DH is an interactive key negotiation protocol.

El Gamal is the key exchange algorithm used in PGP5.  For some reason
PGP Inc insists on calling El Gamal "DH".  El Gamal is a variant of
DH; so it is related.  But El Gamal is not DH, and it is bad
terminology to call it DH.

The only reason I can think that they insist on calling El Gamal "DH"
at every opportunity is that they perhaps think that the name DH
(Diffie-Hellman) is more widely known, and want to give people warm
fuzzies "oh I've heard of that algorithm".

> and DSS is only capable of signing.  Hence, you require one key to
> sign, and one to distribute the session key.

right.

Adam
-- 
Now officially an EAR violation...
Have *you* exported RSA today? --> http://www.dcs.ex.ac.uk/~aba/rsa/

print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<>
)]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`






Thread