Gregory and Ted - Thanks for sharing your ideas. The challenge here runs deep...
At this time, I am measuring the 'effectiveness' of my contribution to Tor though the bandwidth that is used. In 2008, I had over a TBytes of bandwidth that I didn't use. (Purchased as a part of my server hosting service, linode.com). I have re-configured my relay to also accept inbound connections. I feel like this will have a low probability of creating friction between my service provider any myself, and as can be seen in the attached graph is using more bandwidth... So I am able to both contribute bandwidth to the project and avoid the problems associated with being an exit node. I am considering running a separate exit note later - just getting my mind wrapped around all of this... Again, thanks for your comments... Erik On Fri, Jan 2, 2009 at 1:25 PM, Gregory Maxwell <gmaxw...@gmail.com> wrote: > On Fri, Jan 2, 2009 at 11:43 AM, Erik Heidt <erik.he...@artofinfosec.com> > wrote: > [snip] > > - Permitting exit to key informational resources (e.g. wikipedia > services) > > - Permitting exit to top 5 or 10 web mail services (e.g. google mail, > > hotmail, yahoo, etc.) > > And manage to make yourself look like someone doing a targeted MITM > attack on tor users. :( > > The above use cases should be handled by improving the exit enclave > support and convincing these sorts of major sites to run their own > exists. > > If you'd like to cut abuse without looking like an attacker yourself > you're best off limiting your exit to ports which cause few abuse > complaints. IRC and other chat protocols are probably good examples. >
<<attachment: TurnOnEntry at 730.png>>
