Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[liberationtech] Please help us promote Tor and online anonymity

Roger Dingledine arma at mit.edu
Thu May 26 13:28:24 PDT 2011


On Thu, May 26, 2011 at 12:02:48PM -0700, Tom Zhang wrote:
> This is a great initiative of EFF. I'll certainly participate, and promote
> it as best as I can. I do have one question/comment, we know Tor is banned
> in China and Tor bridges are partially banned as well (see
> https://blog.torproject.org/blog/china-blocking-tor-round-two), how
> effectively would this Tor expansion initiative fare in China if China is to
> more aggressively block Tor? I hope you can give us an analysis on this, if
> the conclusion is optimistic, then it'll help the promotion of this
> initiative hugely (otherwise my friend would say "why bother").

Hi Tom,

Let me try several answers on you. Let me know which one(s) you like.

First, China isn't the whole picture. Many thousands of people in Saudi
Arabia use Tor bridges right now, even though for blocking-resistance
they don't need to (Tor works just fine out of the box in SA). Tor
bridges actually serve two roles: a) they let you reach the Tor network
even if the public Tor relays are blocked, and b) they make it harder for
somebody watching you to realize that you're using Tor, because you aren't
connecting to an IP address that's easy to associate with Tor. People
in a variety of countries are using Tor for reason 'b', because they
are concerned about their safety. More bridges will help them.

Second, some bridge distribution strategies are currently broken in
China, but others are working quite well. In part the reason why some
are broken is because we don't have enough bridge addresses, and there
isn't enough churn:
https://blog.torproject.org/blog/strategies-getting-more-bridge-addresses
I bet if we had a thousand new bridge addresses daily, the gmail based
distribution strategy would work a lot better.

We're experimenting behind the scenes with other distribution strategies.
In particular, one of the recent approaches that I'm optimistic about is
distribution via social connections. We're doing some of that on our side
(using a subset of the bridges and people we know), but you can also do it
yourself by unchecking the "Automatically distribute my bridge address"
checkbox in Vidalia and then telling your bridge address line to your
friends in China.

Third, increasing the overall capacity of the Tor network (by running
exit relays or non-exit relays) makes the Tor experience better for
the hundreds of thousands of users (some of whom are in China) who can
reach the Tor network. Over the past year we've tripled the capacity of
the Tor network:
http://metrics.torproject.org/network.html?graph=bandwidth&start=2010-05-26&end=2011-05-26#bandwidth
and we've seen a corresponding improvement in consistency of latency for
Tor traffic. Capacity and latency aren't exactly tied together, because
we're working on protocol improvements to handle congestion better (good)
and because new Tor users show up as it gets better and add load to the
network (bad). From what I can tell the set of people who want to use
Tor is basically unlimited; our job is to figure out how to handle more
of them.

--Roger




More information about the liberationtech mailing list