Re: [freehaven-dev] Who's on First protocol

[Roger Dingledine <arma@mit.edu>]

On Sun, Dec 31, 2000 at 05:41:13AM -0500, seph wrote:
> I have not read the paper yet. but back quite awhile ago, I was trying
> to come up with this. We couldn't come up with a system that allowed
> this to work... there was always something that you could do ahead of
> time. take a snapshot of where the virtual nodes point, etc...
 
The problem with our reasoning was that we assumed all of the virtual
nodes (that is, remailer addresses) would be published. Thus you could
take a snapshot of all of them, and then later realize that A pointed to
B, and from there reconstruct the actual destination of A.

But in the case of wof, B is private. Only the end-node of A knows
what B is, and presumably it tries to forget old ones. Therefore it is
more difficult to "just take a snapshot of the whole system", since you
can't easily get the whole picture, and you don't immediately know which
virtual nodes you might want to investigate further.

--Roger