Re: [tor-talk] Tor (and other nets) probably screwed by Traffic Analysis by now
On 6/2/16, Aymeric Vitte <vitteaymeric@gmail.com> top posted without trimming:
Yes: https://github.com/Ayms/node-Tor#convergence
Let's imagine that one Tor circuit reaches a P2P network (here browsers) and is splitted between different peers (UDP) circuits before reasynching to a relay or end point, then the reconciliation from the source to the end point is quite unlikely
Is there any group / list that is actively researching or developing such networks? Or that wants to?
If I remember right you're doing file caching / serving in the browsers-as-torrent-nodes layer. Yes that asynchronous would seem to break end to end timing. But the 654321012 byte file "into one tor circuit" from user at one end would still count same bytes out to a "reasynching end point" user at the other end. (Or users plural if requested by more than one user). Between otherwise relatively quiet endpoints / users. If they are all shuffling around encrypted bits fulltime, without being driven to do so by end user demand, such that the start and stop of the bytes of the poster and requestor[s] can't be delineated, that looks more generally like the fill traffic.
participants (1)
-
grarpamp