[tor-talk] Contents of PirateBrowser 0.6b

Roger Dingledine arma at mit.edu
Sat Aug 31 04:25:36 UTC 2013

On Fri, Aug 30, 2013 at 04:29:18PM +0000, Matt Pagan wrote:
> # Configured for speed

Just for the record, the three lines here don't help speed much (or
maybe at all).

> ExcludeSingleHopRelays 0

This first line says it's ok to use relays that allow you to make one-hop
circuits. Roughly speaking, there aren't any relays like that in the
network (it's not the default). And even if there were, it wouldn't make
the circuits your client builds any speedier (except I guess through
a second-order effect where you're willing to use relays that all the
other clients are unwilling to use).

> EnforceDistinctSubnets 0

This makes you willing to use more than one relay on a given /16 subnet
in your circuits. So if there are two relays on a single computer, you
might use one as your entry and one as your exit. This could help your
performance, but it sure does reduce your anonymity. (Also, it makes
Pirate Browser distinguishable from normal Tor clients in a really
interesting way -- for example, middle nodes can sometimes distinguish

> AllowSingleHopCircuits 1

This disables a protection that prevents Tor controllers from manually
selecting relays that allow single-hop circuits. Assuming they didn't
also change the controller to do such a thing -- and remember also that
there are roughly no such relays anyway -- it doesn't help speed any
that they disabled it.

> # Exclude countries that might have blocks
> ExcludeExitNodes {dk},{ie},{gb},{nl},{be},{it},{cn},{ir},{fi},{no}

This line probably has the biggest impact on performance. Whether it
improves or reduces performance I cannot say, though. Depends on a lot
of complex variables around Internet topologies.


More information about the tor-talk mailing list