Stay on Top of Enterprise Technology Trends
Get updates impacting your industry from our GigaOm Research Community
Johan Pouwelse is a busy man. The P2P researcher based in the city of Delft, in the Netherlands, is heading up development of the social BitTorrent client Tribler; he’s also deeply involved with the EU’s P2P Next project, which aims to use P2P streaming for an open source, next-generation video delivery infrastructure. And Pouwelse, who’s been tracking the P2P phenomenon over the last decade, has just published along with some of his colleagues an article highlighting some of the key points of his research. It’s a good 21-page read, but here’s the short version: That whole copyright thing ain’t gonna work.
In fact, Pouwelse thinks the existing copyright system could fall apart as early as next year unless significant reforms are put into place. He draws this conclusion from an analysis of not only movie file-sharing, but activity on social networks like Facebook and streaming video sites like YouTube. All of these platforms are prime examples of user-based collaboration, or peer production, as Pouwelse likes to call it. These forms of peer production are not only getting more and more popular, but also increasingly sophisticated, to a point where they poise a significant challenge to our established system of content production and monetization.
OK, I’ll admit it. Part of me was skeptical when I browsed through his article, entitled “Pirates and Samaritans: a Decade of Measurements on Peer Production and their Implications for Net Neutrality and Copyright,” for the first time. Combine some of the most popular social platforms in one article, throw a few buzz words in the mix, and you got yourself a theory that touches everything but explains nothing. Or so I thought. But once I started to read a little more closely, I realized that Pouwelse and his colleagues are actually onto something.
Take KaZaA. The file-sharing application once attracted millions of users, but struggled with the fact that most were neither willing nor able to share as much as they downloaded. Pouwelse and his colleagues analyzed file-sharing behavior on KaZaA in 2003, and found that almost 40 percent of users contributed barely any upload bandwidth at all, while another 25 percent contributed less than half of what they consumed. KaZaA tried to solve this problem by introducing ratios, but since users were practically anonymous, there was no real incentive to behave well.
Things have changed a lot since then, as YouTube shows. While the video-sharing site isn’t technically a P2P system, its users are nonetheless sharing content, and they’re utilizing networks of friends to do so. Pouwelse’s team has crawled more than 5 million of YouTube’s user pages in the last two years, and discovered a tight social fabric in the process. “(T)he concept of a friend is taken to an extreme,” the article explains: “The top 10 YouTube users with the highest amount of friends together boast 72,866 ‘friends’.” Yet most YouTube users only have a few friends at a time. Only around one hundred of the users analyzed for the article had more than 1,000 friends.
So what does all this have to do with the impending doom of copyright? A lot, at least according to Pouwelse. Collaborative web sites and sharing platforms like YouTube are getting more popular and more social, leading to more shared content and better mechanisms to filter though this content. P2P networks, on the other hand, are getting more robust and increasingly capable of adopting some of these social networking and filtering mechanisms.
At the same time, they’re getting more secure: “By 2010 darknets should be able to offer the same performance as traditional P2P software by exploiting social networking,” the article reads, referring to networks that allow file trading whithout revealing the identity of its participants to outside entities. Just think what would happen if those 72,866 YouTube friends were able to share Hollywood movies within a P2P network that’s as easy to use as YouTube but untraceable by Hollywood. Pouwelse and his colleagues think it’s going to happen within the next two years.
Granted, there are plenty of reasons to doubt whether this really means that copyright as we know it will completely fall apart, and there are other parts of the article that are lost on me, like the reference to net neutrality. But it’s nonetheless an interesting read, if only for one fundamental thought: File sharing, as the article describes it, is a social behavior, and a solution to all the related copyright issues won’t be found until this social aspect is addressed.