Allow me to combine some quotes from this recent thread alone...
I've had an idea for a while for a killer service for...
I'd be very much interested to see it in reality. I guess the delays will be more then acceptable.
it would be cool to have a more general protocol for P2P use through hidden services.
Hidden Services as a globally anonymous NAT'd networked is an amazing opportunity for this type of application.
It seems pretty clear we are at the point where there are current and growing ideas and demand for all manner of anonymous P2P services. In part driven by global news, and thus global awareness, that governments, business and others have questionable regard for privacy, etc. And if consumers continue to demand and drive openness into products, there will soon come a day when open, BSD/Linux capable, pad devices (PC + wifi + cell) the size of a DVD case are common. Demand, compounded by utility. Regardless of the capacity for the sort of projects listed below to handle it, people WILL build these services. Some of them are out there today, and being largely unknown, they are performing well enough. More users will try them and adopt them based on need, utility and performance. The only question is how will the networks perform at each stage of popularity. Given that these services are surely coming in force... and from directions that see these networks more as a raw transport than say, primarily for the purposes outlined on their respective web pages... it seems the usual echo of "we're a nice project, don't do that, too much load" may not be useful. This should not imply fault, but merely suggest that it's dated and hopeless. And since the networks are anonymous, countermeasures would affect all users matching the metric, even those using it for said purposes. So maybe there's room for more focused/funded research into the actual impacts of different traffic models (leading to proper resource giveback by apps/users/groups), development improvement in the networks to handle the traffic, even up to creating an entirely new anonymous network optimized for P2P ...
It is not unheard of for apps that work well and do something cool to suddently have 1M+ users
... before this certainty happens.
which isn't to say you shouldn't build this.
What should I do then to have anonymous and secure decentralized chat, file sharing and the like on Tor?
We just need to solve the scalability problem so we do not, as Robert put it "thrash the hidden services directory".
Do what is needed to go forth and conquer :) P2P services should look at the current strong anonymization projects that could provide transport. The ones I know of are: http://www.i2p2.de/ https://code.google.com/p/phantom/ https://www.torproject.org/ More links are welcome. People should not make the mistake of assuming they can freely use any network without negative impact. Including fatal impact for their own service. They should offer sufficient resources back to the network. Contrary to popular belief, P2P is not a free ride. Relay, directory and index is expensive and the cost is bandwidth, CPU and storage. _______________________________________________ tor-talk mailing list tor-talk@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk ----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE