[tor-talk] Verifying node operator identity (was: New paper : Users Get Routed: Traffic Correlation on Tor by Realistic Adversaries)

Aaron aagbsn at torproject.org
Thu Oct 17 09:47:53 UTC 2013

On Thu, Oct 17, 2013 at 7:52 AM, Eugen Leitl <eugen at leitl.org> wrote:

> On Thu, Oct 17, 2013 at 08:18:59AM +0200, Felix Eckhofer wrote:
> > tried very hard to find any suggestion on how this might work in
> > your mail to no avail. Are you actually suggesting "extensive
> > personal interviews, background checks, giving polygraph tests,
> > injecting sodium pentathol to those wanting to run nodes" and expect
> > some form of serious feedback?
> It would be good if operators of fast exits were personally known,
> and had a history within the community. This still doesn't prevent
> deep cover moles or people turned through threats and/or bribes,
> and as soon as traffic exits it will be scrutinized, sliced and
> diced ten time to Sunday, anyway. But it would be good, yes.

Security based on trusting the exit operators won't work; even the most
honorable operators are subject to upstream taps or could have their exits
broken into and backdoored. Disincentivsing people from helping the Tor
network by suggesting that in order to help you have to give up personal
privacy seems more likely to hurt than help.

A different approach would be figuring out how to increase the diversity
and numbers of fast exits, perhaps by strengthening protections for running
exits in countries where that isn't the case, and educating lawmakers and
enforcers so that perhaps some day our advice won't be "Don't run an exit
at home" [1] but look more like that of the Open Wireless project [2].


[1] https://blog.torproject.org/running-exit-node
[2] https://openwireless.org/

> tor-talk mailing list - tor-talk at lists.torproject.org
> To unsubscribe or change other settings go to
> https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk

More information about the tor-talk mailing list