[liberationtech] Anonymity Smackdown: NSA vs. Tor

[Jonathan Wilkes]

On 08/07/2013 03:46 PM, Guido Witmond wrote:
> On 07-08-13 20:47, Jonathan Wilkes wrote:
>
>> Anyone outside of the VPS and the attacker cannot know whether that
>> relay/exit node has been patched.  Now just work with the NSA's
>> equivalent in the country of the exit node to make sure the VPS
>> remains unaware of any shenanigans (and why wouldn't they?).  If they
>> do see something weird happening from the patch and make some noise
>> about it then just gag them as the Guardian reports show they are
>> quite good at doing.
> No need to gag, thank the VPS provider publicly for spotting the hacked
> node.

Good point.

>
> Then run a Freenet/bittorrent/I2P/etc node on it to hide the spooks
> traffic and sniff that with their fiber taps.
>
>
>> Roger-- how exactly would you check to make sure something like this
>> scenario isn't happening?
> Hmm, That would be easy. Place some false flag mails about terrorist
> attacks and check for raised alerts... :-)

Wouldn't that be difficult?  When cross-referenced with the greater
social graph built from all available sources those false flag mails 
would look like
stumps.  They wouldn't connect up with cellphone metadata, social network
activity, people under targeted surveillance, etc.


Guido.

PS. The best thing is to discourage plaintext protocols. Encrypt
everything. Then the spooks have *only* the metadata if they manage to
trace the Tor paths end to end.


I posted an idea on the Bitmessage forum about putting a feature in that
queues up a message each day to be sent at 0 UTC (or else it sends garbage
to a random address), in order to make it harder to even get metadata.
Sounded like it wasn't practical, though.

I don't know enough to implement something that is practical, but I hope 
the people
who have that kind of expertise revisit the feasibility of building such 
metadata-
snooping resistant networks.  Assumptions about what kinds of inconveniences
people are willing to put up with (like latency) have probably changed 
quite a
bit, even in the past few months.

-Jonathan

> --
> Liberationtech list is public and archives are searchable on Google. Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at companys at stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
>