[tor-dev] Using TorPS to estimate chances of all-TAP paths

Roger Dingledine arma at mit.edu
Thu Sep 19 10:03:09 UTC 2013

Hi Aaron, Sathya,

I have another use for your fine path simulator. :)

See https://trac.torproject.org/projects/tor/ticket/9777

The first question is: for a recent consensus, what's the chance that
your middle hop and your exit hop will both be running
or earlier?

The extra credit for question one is to generate the answers to question
one for each of the past n consensuses (where n is, say, the past month)
so somebody can graph them.

The second question is: if we change path selection to try again if
we choose such a path, what does that do to our anonymity, for various
definitions of anonymity like the ones in the upcoming CCS paper?

The extra credit for question two, if you haven't guessed it already,
is to get some data points for question two from the past as well.
I'm guessing getting 24*30 of them will be no fun, so whatever isn't no
fun is fine.

What can I do to help make it happen? If the answers are "actually
not all that high" and "oh hey those curves aren't all that different"
respectively, I want to put #9777 into 0.2.4 and enable it. If those
aren't the answers, we'll need to sort out a better plan.


More information about the tor-dev mailing list