said by jester121:In a nutshell, and without excuciating detail because I'm not a mathemetician, P2P software works by creating an MD5 hash based on the contents of the file. There's a very high degree of certainty that files with identical hashes are the same, regardless of what the file is called on each user's storage medium.
They're "assuming" that if they find files on a peer that match the hash of their known copyrighted file, the contents of that file is the same as the content of their "control" and they can nab you for distributing.
I have no idea what the chances are of two files of identical size having different contents but the same MD5 hash. I'm thinking it's infintessimally small, like smaller than DNA-match-certainty type numbers. I do know if you download a movie, use editing software to clip a second off the front or the back, and re-hash it, you'll get a completely different result.
What you are saying is correct so far. The problem with your analysis is that the hash is not compared with a control copy but with the master copy that is being seeded/sent. Someone creates a copy and seeds it. This copy's hash is listed in the .torrent file that is made available from the torrent server. As clients use the .torrent file and the server, the master .torrent file copy is updated with the list of IPs of who has parts of the file. To see the contents of the file, the monitoring organization needs to download a full copy. As they do that the client tracks who is supplying each piece. This list will thus list who has parts of the file. Anyone listed as a leach has part of the file while being listed as a seed means you've downloaded the full file. So long as you sent a piece to the monitoring organization's client, you have at least part of the file. Once all the pieces have been captured a hash is made of the downloaded copy and compared with the one in the .torrent file. A match means that the copy is an accurate duplicate of the master copy being seeded.
