On Sun, 10 Mar 2013 11:32:20 -0700 Yosem Companys <[email protected]> wrote:
> http://hackertarget.com/tor-exit-node-visualization/ > > Tor Exit Nodes Located and Mapped Tor includes a Network Map which also maps nodes across a Mercator projection of the globe. I don't understand this fetish with Google mapping everything (why not openstreetmap? why not doing lat/long coordinate mapping on your own map?). However, it apparently exists. Ok, moving on. The usage of blutmagie is an interesting choice, when https://metrics.torproject.org/ and https://compass.torproject.org/ and https://atlas.torproject.org/ all exist with gobs more data, including the raw source data on which all of those sites are built. Freegeoip.net uses the same MaxMind GeoList database we do. Google has it's own GeoIP database as well. I wonder which one is most accurate. What's more valuable to me is the list of exit relays by ISP. Intuitively, it makes sense. Cheap, well-connected server providers are going to be attractive to those running Tor relays (exits or not). I take this list to mean, "this is the list of ISPs who care about Internet freedom". It sure seems small. The other implication here is that Tor exits can be monitored. Yes, we know. We've been saying this for a decade. Here's one blog post about it, https://blog.torproject.org/blog/plaintext-over-tor-still-plaintext. It's part of the reason the EFF and Tor write HTTPS Everywhere for Chrome and Firefox, https://www.eff.org/https-everywhere. Your ISP can watch your traffic too, and inject ads or redirect DNS. In fact, there's a billion dollar market for traffic management at ISPs. I'm assuming this means the providers are looking to manage Tor traffic as well. For me, the visualization is a cool map, a product pitch for hackertarget llc, and just raises some implications for operational security. Larger questions it raises in my mind are about AS-aware routing, what does it mean to have a large concentration of exit relays in pro-freedom ISPs, and what about legal jurisdictions and MLATs? There is some on-going work on the AS-awareness question, see http://freehaven.net/anonbib/bibtex.html#DBLP:conf:ccs:EdmanS09 and http://freehaven.net/anonbib/bibtex.html#oakland2012-lastor and http://freehaven.net/anonbib/bibtex.html#ndss13-lira As for some measurement of anonymity and risk modulo concentration of exit relays? Who knows. Sounds like a fine project. And I know of at least one group working on the MLAT and legal arbitrage question as it relates to tor circuits and relays. I look forward to more analysis and proposed research theories to improve the Tor network in the future. -- Andrew http://tpo.is/contact pgp 0x6B4D6475 -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at [email protected] or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
