Re: A timestamp attack on Babel's anonymity

[George Danezis <gd@theory.lcs.mit.edu>]

I agree Roger,

The issue of the technicalities of implementing mix networks have never 
really been researched in detail. As soon as mix technology got in the 
hand of the cryptologist it all became zero-knowledge proofs without 
realizing that the systems cannot be implemented for engineering reasons or 
are vulnerable for practical reasons (timestamps, etc).

George

On Tue, 7 May 2002, Roger Dingledine wrote:

> I went looking in the Babel paper to see what its replay policy is.
> 
> |Section 8.6. Time Synchronization & Replay Detection
> |
> |As mentioned in Section 3.2.2, each layer of the onion created by Alice
> |includes a time stamp. The value of the time stamp, referred to as \Theta,
> |is the number of seconds elapsed in seconds since January 1, 1970 GMT,
> |to the moment of message composition by the sender.
> 
> I can only assume that each layer of the onion has the same time stamp.
> 
> There probably aren't all that many people who compose a message on the
> same second.
> 
> --Roger
>