Alternative BitTorrent Protocol
3 comments
If you had everything go through two or more "feeders" then it would be harder for the enforcers to control the whole chain. Also, what if everyone just appeared to be a feeder. So a feeder doesn't know if the message it's receiving originated from the sender or is just being passed on by the sender. I'd imagine this would result in a pretty slow protocol. If I'm not mistaken this is kinda how Tor works though.
Yes, I'm beginning to see the problems with this scheme, particularly with speed and scope. Looking at Tor, I see that it is, essentially, the same concept.
You should post this to http://www.reddit.com/r/darknetplan, too. That subreddit is very interested in ideas like this.
btguard.com is pretty much this. But you need to pay for it, because getting decent speed requires some servers.
I have an idea for a new BitTorrent-esque protocol that might have a chance of circumventing legal issues that commonly arise with file-sharing.
Essentially, where BitTorrent contains Seeders and Leechers, I would add another class called, I dunno, let's say "Feeders," that act as a proxy between the other two. When a Leecher requests a file from a known Seeder, the data is first sent through a (random) Feeder as a proxy, which records no data on either the file or the Leecher. Since, the Feeder is not actually storing the file and doesn't record the Leecher's IP, it shouldn't be illegal for the Feeder to act as a proxy. As of now, I use BTGuard as a proxy for my BitTorrent connections, which does essentially the same thing: proxies my BT data and keeps no records. This is how a "pirate" would be able to use BTGuard to anonymously share files.
One flaw in this scheme that may turn up is Copyright-Infringement Scouts could act as a Feeder, picking up IP addresses that turn up (only assuming they used a packet sniffer to detect IPs, because the Protocol itself would not do so), but this could potentially be fixed by using 2 Feeders per Seed, which further obfuscates the Leecher.
The legal burden might then fall on Seeder for having Copyrighted material on the open web, but as I understand it now, currently the legal burden falls on the Leechers when they download and are caught by CI-Scouts, which is what I am trying to fix.
Hopefully I'm not grasping at straws. I am really keen on starting this, assuming that such legal issues could be avoided. I am currently in my second semester of Networking classes and I am confident that, given a certain amount of time, I could complete such a protocol.
Before I look further into doing this, I wanted to run it by some other hackers, see what you guys think. Any input would be appreciated.