Quite simply, my argument is that most ISPs don't take even the most basic cybersecurity precautions (spoofing protection) at the user end. Therefore it is reasonable to assume that for them to install significant traffic analysis on every user would be quite expensive. Now, the state can require that they do very expensive things and increase their prices accordingly, but this is a nontrivial cost.
My french isn't really good enough to get to grips with that paper; what exactly DID they say about traffic flow analysis? It *looks* like they said it wasn't feasible except on a very localised level with current technology... right? The obvious exception to all this is BT, which is apparently installing web blacklisting systems to block child porn. Of course, how long it remains just the self-regulating-ISP-thingy's list of known child porn sites, we will see; I'd be very surprised if it didn't grow to cover all illegal content of any kind, including libellous material and so on... On Wed, Oct 12, 2005 at 02:14:50PM -0400, jrandom at i2p.net wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > > Simple economics, freenet 0.5 hasn't shown that its worth blocking yet. > > > > ... Except that they HAVE blocked it. > > You're right, I should rephrase it - Freenet 0.5 hasn't shown that its worth > spending a substantial amount of time or resources blocking yet. 0.5 can be > fixed with an IV or a whole slew of different techniques that would defeat > the most basic blocking in place. At that point, it becomes a question of > what the next cheapest way to block it is. > > > > $1k/ISP plus N smart-coder-hours plus J support hours plus legislation is > > > more expensive than blacklisting a website plus someone running a sniffer, > > > browsing Freenet's source, or looking at one of those commercially > > > available network identification tools which already can identify Freenet. > > > > They can't identify darknet traffic that easily. And $1000/ISP? If J is > > non-trivial, the overall cost will be much higher than that. Just for > > the hardware/software changes, $1000 is absurdly low. > > Ok, you can keep replying back with suggestions that its not possible without > any basis in fact, but I've already shown that its not just possible, but the > infrastructure necessary is already economically feasible and infact in use > with at least some ISPs, even without a state mandate. > > If you're going to make the positive affirmation that its not economically > feasible to identify Freenet/dark's traffic, you'll have to back up that > statement (beyond a report focused on telling copyrighted materials from > those that aren't). > > Why should you bother? Because, as you say, the whole point of Freenet/dark > is to operate in places where its not feasible to identify the traffic, since > Freenet/light would not offer the necessary anonymity for their users. > > I'm not expecting an answer today or tomorrow, but such a feasibility analysis > should be on the critical path of evaluating whether the Freenet/dark effort > is worthwhile for more than academic purposes. > > =jr > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.1 (GNU/Linux) > > iD8DBQFDTVDDWYfZ3rPnHH0RAqs+AJ4sJgaK3AEwEIOjlNlJ0Hts9IW5tACfVGz3 > XyLrAo8JJ82kdXYnLa/OjH8= > =Zcsj > -----END PGP SIGNATURE----- > _______________________________________________ > Tech mailing list > Tech at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/tech -- Matthew J Toseland - toad at amphibian.dyndns.org Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/tech/attachments/20051012/5e13c4a8/attachment.pgp>
