[Cryptography] Are Tor hidden services really hidden?

Peter Todd pete at petertodd.org
Sat Feb 15 13:52:14 EST 2014

On Fri, Feb 14, 2014 at 11:50:01AM -0500, Tom Ritter wrote:
> On 14 February 2014 11:19, Phillip Hallam-Baker <hallam at gmail.com> wrote:
> > Again this is raising the cost of the attack, not preventing the really
> > determined attacker.
> Agreed, wholly.
> > It could be that they are all being found because they are making stupid
> > mistakes like sending email which has to go outside the Tor system becauseof
> > the spam controls. But it wouldn't surprise me if we later find that there
> > are tens of thousands of NSA/GCHQ run nodes.
> I'll nit and say we only have 5K nodes, including 1K exits and 2K
> guards but it's possible.  I proposed (and intended to do before I got
> sidetracked) that we just go through the exit probability percents and
> tick off which nodes we believe are run by trustworthy people, and
> just see what percentage we get to.  I think it will be less than 50%,
> but greater than 25%.

A reasonable "plausibile deniability" approach for anyone wanting to run
a hidden service may very well be to run their own relay/guard nodes and
use those nodes in the routing path. Of course, you'll want to make sure
the security, including physical security, of those nodes is good enough
that they are trustworthy.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 685 bytes
Desc: Digital signature
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20140215/5b149e39/attachment.pgp>

More information about the cryptography mailing list