From: jamesd@echeque.com
Date: Sat, 30 Mar 1996 15:11:47 +0800
To: perry@piermont.com
Subject: Re: Edited Edupage, 24 March 1996
Message-ID: <199603291625.IAA09564@dns2.noc.best.net>
MIME-Version: 1.0
Content-Type: text/plain


Perry E. Metzger wrote:
> 2) I strongly hope that Netscape tries to move the product towards
>    standards based mechanisms like the IETF's RTP protocol, which are
>    in widespread use,

Unfortunately RTP is not a crypto protocol, and does not have a 
standardized encrypted form.  

Therefore any encrypted protocol is necessarily proprietary and
non standard, unless Phill Zimmerman has published a standard.

If Netscape creates a standard for encrypting RTP, and publishes it,
that will be a move towards a standard, not a move away from a standard.

One mechanism for encrypting RTP would be to construct a shared secret
key by DH exchange, or Rabin if one wished to dodge patents, construct
a cryptographically strong pseudo random data stream from the key, using
Ron's code, and for each RTP packet, encrypt using a block from that 
data stream as the packet key.
 ---------------------------------------------------------------------
              				|  
We have the right to defend ourselves	|   http://www.jim.com/jamesd/
and our property, because of the kind	|  
of animals that we are. True law	|   James A. Donald
derives from this right, not from the	|  
arbitrary power of the state.		|   jamesd@echeque.com