-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 toad at amphibian.dyndns.org wrote: > I apologize for provoking a flamewar, but the article is utterly untrue > and a personal attack. This may not be the author's intention, but it is > the fact of the matter.
Is this a flamewar? Oh well. Whatever. As an ex-freenet user, I've moved to I2P for some reasons: speed, usability, the fact it's dynamic rather than static, the feeling it's more versatile overall. This was back when freenet was 0.5-something. I reckon I2P wouldn't last more than few seconds in China. Minutes, perhaps, but that's not the point. Harvesting makes things very hard in countries where it's forbidden. I do recognize the darknet idea is a good thing. But, I have a few doubts regarding what you say about darknets. First, why would it be a slow process to catch the darknet users? once you have a mole in the network, you (governmet agency) can rather easily check on traffic flow between users and outline the network. You just need the traffic logs for a week or so. Also, once you catch a node, you can easily see the direct peers and repeat the (abduction) process. What mechanism do people have to remove the proof that they used the darknet? Unless you implement the darknet as a hidden feature of a modified pre-existing P2P program (say eMule), so that its trafic is submerged between the other requests, you're open to traffic analisys. Secondly, why do you think the network would become a small world? (a FAQ or paper link would be enough) Somehow, I can hardly imagine why someone would build loops in the chain. I'm more prone of thinking about a rather random-looking tree structure. Expecially in situations like China, adding a new node is risky. Would YOU talk about adding someone who says he's already in the network? He could be lying. He could be dangerous. Unless of course you're talking about each chinese connects to someone outside China for the darknet, in which case I suspect the gentle users on the our side ot the wall would quickly get blacklisted, once again cutting out chinese users (and paying them a visit for connecting to a forbidden address) Lastly, speed. While I2P can easily balance traffic to unloaded nodes, a darknet cannot do that, for the simple fact that its possible connections are in a much smaller number. And, it's easier to saturate a darknet because its connections are arbitrary and not optimal in some performance metric. I believe a caching scheme can mitigate the problem, but not solve it. In fact it's plain possible that a number of similarly-interested users are connected in very different places of the network. Last words, I do think that freenet 0.5 was rather a failure. I also think 0.7 has some interesting possibilities, and as such it's not dead meat in its cradle. A scalable darknet IS interesting stuff. But, I2P over freenet, not yet. This is some very future talk. So spoke I. And I might be wrong :D -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFDQ8nqMiWEUf7YMGERAlurAKDOevxQv8BWZ00MKEZfUl87kM/pzgCgk5Bw lZPz9t1+AHs3N2E1LFS4HZE= =8nnS -----END PGP SIGNATURE-----
