[tor-talk] New paper : Users Get Routed: Traffic Correlation on Tor by Realistic Adversaries

[Eugen Leitl]

----- Forwarded message from Joe Btfsplk <joebtfsplk at gmx.com> -----

Date: Wed, 16 Oct 2013 19:42:41 -0500
From: Joe Btfsplk <joebtfsplk at gmx.com>
To: tor-talk at lists.torproject.org
Subject: Re: [tor-talk] New paper : Users Get Routed: Traffic Correlation on Tor by Realistic Adversaries
Message-ID: <525F3281.9030703 at gmx.com>
User-Agent: Mozilla/5.0 (Windows NT 6.0; WOW64; rv:17.0) Gecko/20130801 Thunderbird/17.0.8
Reply-To: tor-talk at lists.torproject.org

On 10/16/2013 4:50 PM, Roger Dingledine wrote:
> On Sun, Sep 01, 2013 at 10:10:56PM -0400, Roger Dingledine wrote:
>> 
>> Yep. They're part of the Tor research community. I have plans for writing
>> a blog post about the paper, to explain what it means, what it doesn't
>> mean, what we should do about it, and what research questions remain
>> open.
> Here it is:
> 
> https://blog.torproject.org/blog/improving-tors-anonymity-changing-guard-parameters
> 
> --Roger
I read the paper - good job. Some of it will be over the heads of
some, but that's unavoidable unless make it 10+ pages, in newbie
language, then few would read it all, so...
I'm not bashing Tor here, so leave your pitchforks in the barn. Just
asking questions, making observations that may / may not have an
answer or even be useful.

One thing jumps out, Tor doesn't know for sure who's running Guard or
exit nodes - & can't unless they start doing (regular, repeated)
extensive personal interviews, background checks, giving polygraph
tests, injecting sodium pentathol  to those wanting to run nodes.  I
guess more so for Guards.

Since apparently now LEAs from (some) countries are teaming up,
sharing info, etc., seems possible the problem of LEAs (or any
adversaries) running a higher % of nodes could get worse, not better.
If adversary nodes as a % of all nodes doesn't increase (new good guy
nodes keeps up w/ increase of adversarial ones), then overall risk
hasn't changed.  But how can Tor (or any group) determine the risk if
they have no reasonably reliable way to determine the REAL intention /
identity of node operators (spies infiltrating Tor Network)?

Governments, crooks have proven themselves VERY resourceful over
decades, or 100's of yrs.  The U.S., let alone other industrialized
nations partnering together, has a lot more manpower, resources &
money than Tor Project.  I don't think we can out spend "them," for
setting up nodes.  How many full / part time programmers or "idea
people" does Tor have (as good as they are) VS. one agency of one
industrialized nation?

Is there any way - in the future, that Tor could run a much larger %
of nodes or at least, instead of constantly trying to figure how to
"beat / drastically improve the odds" that an adversary won't
accidentally control the entry / exit nodes on circuits?  Perhaps a
noble, but losing game, if gov'ts band together & decide Tor, or the
entire internet, IS worth serious monitoring.  Perhaps reasonable
anonymity on a world wide party line is too ambitious? (Those that
don't know what a "telephone party line" was, can "Startpage it."
[stop saying "Google it"] :)

What about somehow getting a better handle on who actually runs the
nodes?  With its current policies & design, Tor is in a very tough
position to "ensure quality" (anonymity).  Tor isn't supposed to see
any real data on the network - for one, so they can't be forced to
give anything up (again, noble), but that prevents some (a lot of?)
capability for quality control.  No company would / could handle its
own security that way.  It's a Catch 22 situation for Tor, because of
legal threats that many gov'ts impose, that many corporations don't
face.  And if they had some REALLY secret stuff to send abroad, they'd
fly it in their own jet.

What about a COMPLETELY different approach, rather than trying to
develop methods to "beat the odds," *ad infinitum,* against what COULD
become an ever increasingly larger PERCENTAGE of gov't / adversary run
nodes?  Surely, it'd be worthwhile to look way down the road & see
where Gov'ts / LEAs may be going w/ this & whether they can be
"bested," by following the same course that Tor is on (even with
improvements along the way)?  I have no idea - I'm just saying,
sometimes the only way businesses, technologies, gov'ts survive &
thrive is to completely change course.   For all of history, gov'ts
have gone to GREAT lengths to spy on citizens & adversaries & have
often done pretty well at it.

Well liked corporations can often be as secretive as they want -
they're "protecting corporate data & assets."  Tor is looked at in
part (*by gov'ts & LEAs*), as a tool for terrorists, criminals - of
all sorts.  They couldn't care less if honest people, whistle blowers
swim near schools of criminals & terrorists, whether some will get
caught in the same net.  Maybe, like Corporations that get away w/
figurative murder, Tor Project should start contributing heavily to
key political figures, to ensure they'll "be left alone?"  :D

You laugh, but that's exactly why big business, who by current
STATUTES, break JUST AS MANY OR MORE laws, as Gov'ts / LEAs *ASSUME*
that Tor users do?   Big Business is left alone & entities like Tor
are on the hit list.
-- 
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

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://ativel.com http://postbiota.org
AC894EC5: 38A5 5F46 A4FF 59B8 336B  47EE F46E 3489 AC89 4EC5