A central theme of Walled Culture the book (free digital versions available) and this blog is that the copyright industry is never satisfied. Now matter how long the term of copyright, publishers a…
Torrents are already very hard to block. You don’t actually need a tracker, because all modern torrent clients support DHT (distributed hash table). You only need some way to get the initial hash for a torrent, so that’s where trackers are still useful, but once you’re connected to the swarm, you can only be blocked if the entire swarm is blocked.
Tracking though… It’s too easy to get IP addresses for the entire swarm and I don’t see how you could ever fix that. Tor doesn’t really solve that issue either, it just moves it to places where you won’t get in legal trouble or to people who don’t mind getting in legal trouble, a bit like VPN providers.
I wonder what the legality would be of seeding the binary difference between the wikipedia.zim and a copy of the wild robot. But I digress… We could probably bolt on something I2P like to torrent, have everything pass through multiple nodes. I fear the best we could ever work out would be plausible deniability.
Torrents are already very hard to block. You don’t actually need a tracker, because all modern torrent clients support DHT (distributed hash table). You only need some way to get the initial hash for a torrent, so that’s where trackers are still useful, but once you’re connected to the swarm, you can only be blocked if the entire swarm is blocked.
Tracking though… It’s too easy to get IP addresses for the entire swarm and I don’t see how you could ever fix that. Tor doesn’t really solve that issue either, it just moves it to places where you won’t get in legal trouble or to people who don’t mind getting in legal trouble, a bit like VPN providers.
I wonder what the legality would be of seeding the binary difference between the wikipedia.zim and a copy of the wild robot. But I digress… We could probably bolt on something I2P like to torrent, have everything pass through multiple nodes. I fear the best we could ever work out would be plausible deniability.