1997-01-21 - Re: Newt’s phone calls

Header Data

From: “Phillip M. Hallam-Baker” <hallam@ai.mit.edu>
To: “‘aba@dcs.ex.ac.uk>
Message Hash: 8287ad0ff163cf4f8d57eebdbcab850a5fcdd437dcb1988777a97dc919e50bed
Message ID: <01BC0792.3DE1F750@crecy.ai.mit.edu>
Reply To: N/A
UTC Datetime: 1997-01-21 16:59:56 UTC
Raw Date: Tue, 21 Jan 1997 08:59:56 -0800 (PST)

Raw message

From: "Phillip M. Hallam-Baker" <hallam@ai.mit.edu>
Date: Tue, 21 Jan 1997 08:59:56 -0800 (PST)
To: "'aba@dcs.ex.ac.uk>
Subject: Re: Newt's phone calls
Message-ID: <01BC0792.3DE1F750@crecy.ai.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain

Adam Back <aba@dcs.ex.ac.uk> wrote in article <5bp18k$1cc@life.ai.mit.edu>...

> GSM encrypts only the links to the station - the traffic goes in the
> clear through the station.  Plus A5 (crypto algorithm used in GSM) is
> weak, 40 bits of effective key space.
> It could be worse to have poor crypto, than no crypto,

I disagree for two reasons, first there is a big difference between having
poor locks and no locks. Most locks can be picked by an expert, they are
effective against many theifs however.

Second if everyone in the world was using 40 bit email encryption it
would prevent most of the "promiscuous" interception of communications.

The danger in weak crypto is thinking that it is strong crypto. GSM is
weak crypto but stops the type of snooping the Martins engaged in.
If you know not to talk about something secret on one then low
crypto is better than having a signal anyone can pick up on a 
device from radio shack.