from the don't-look-now,-you're-being-watched dept
The BitTorrent protocol is an extremely efficient way of moving files around the Internet, especially big ones. That makes it highly popular with those seeking to download unauthorized copies of music and films, for example. But the clever approach that enables BitTorrent to do that, which involves downloading fragments of a file from a shifting swarm of fellow peers holding some or all of it, is also a weakness from these users' point of view: it means that downloads take place in public, rather than as a private transaction from a client to a server (as with cyber lockers.)
Fascinating new research entitled "The Unbearable Lightness of Monitoring: Direct Monitoring in BitTorrent" (pdf), from researchers at the University of Birmingham, explores just how public. It seeks to quantify how many peers in a swarm are actually being run by companies monitoring unauthorized downloads, and how long it takes for them to detect such activity:
From our experiments, we derived a number of interesting properties of monitoring, as it is currently performed: e.g., that monitoring is prevalent for popular content (i.e., the most popular torrents on The Pirate Bay) but absent for less popular content, and that peers sharing popular content are likely to be monitored within three hours of joining a swarm.
Many BitTorrent users are aware that such monitoring is going on, and try to avoid detection by using some kind of blocklist designed to catch peers being run by companies offering digital forensics. But if the current research is accurate, those lists have big gaps: of the peers the researchers identified as likely to be run by monitoring companies, only 69% were found on the blocklists they checked. In other words, there was nearly a one in three chance that they would not be blocked, and would therefore be able to monitor unauthorized downloads.
Despite this, one of the study's authors, Tom Chothia, told New Scientist that file sharers needn't be worried too much by the revelation they are being monitored -- at least, not yet:
"All the monitors connected to file sharers believed to be sharing illegal content. However, they did not actually collect any of the files being shared. So it is questionable whether the observed evidence of file-sharing would stand up in court."
More cautious users of BitTorrent may find other ways to avoid detection, using VPNs or proxies. But as Tim Lee points out, the main effect of this new analysis will probably just be an escalation of the long-standing arms race between file-sharers and copyright enforcers:
Users will presumably take advantage of the new monitor-detection techniques identified by Chothia et al to produce more accurate blocklists. Monitoring firms may respond by tweaking their monitoring clients to behave more like real clients, and by more frequently changing the subnets they use for monitoring.
And so the game of cat-and-mouse will continue.