On 31 Dec 2000, Steven Hazel wrote:
> Scott Gregory Miller <[EMAIL PROTECTED]> writes:
>
> > The protocol should completely defeat the Man in the Middle attack
> > as well.
>
> Your protocol doesn't deal with MITM at all. It assumes that Alice
> already knows Bob's public key, and MITM only works in cases where
> both Alice and Bob fail to actually receive the other's public key.
> If we assume that Alice has been fooled, and has Mallory's public key
> instead of Bob's, your protocol does nothing to foil MITM. There is
> no known way to foil MITM without a prior shared secret.
Yes, thats correct. But remember, Freenet is a bit unique in that your
allowed to generate identities whenever you want, so you can't really
eliminate Mallory if he wants to be a node. The protocol prevents MITM in
a very specific way. It prevents Alice from ever connecting to a person
she wasn't expecting. The current protocol does not do this.
The second half of the puzzle is in the way node public keys are
distributed. They only occur in the key exchange (Alice->Bob), the
StoreData messages of requests, and in the keyspace probe.
We had a length argument about this before. Freenet doesnt really deal
with the classical MITM attack because of the inherent lack of trust built
into the system. The most you can ever do (without breaking Freenet) is
prevent someone from connecting to Mallory when they thought it was Bob.
This protocol does that.
_______________________________________________
Freenet-dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/mailman/listinfo/freenet-dev
