Phil Gyford

It’s interesting to compare Friendster, a Six Degrees-style site for building social groups, and FOAF, an open way of defining your own social network. Both do pretty similar things but with some crucial differences. Differences that resulted in an explosion of Friendster links and invites across the net in the space of a few days while FOAF languishes in the world of Perl Mongers and the technically curious. So how could FOAF learn from this…

It’s easy to see why Friendster has been an immediate hit. Just like Six Degrees was, it’s an intriguing idea that’s easy to grasp and the application is designed to be viral; it’s almost impossible to play with the site without generating email invites to people you know. As commercial sites go Friendster is simply designed and the ability to see photos of all your friends (and their friends) makes it instantly more human than a simple list of names would be. It’s easy to use and there are near-instant results that draw you in further.

On the other hand FOAF is a geek’s toy. It is underlying technology that’s still having front ends built around it and as such isn’t something people are going to play with unless they enjoy fiddling with techology for its own sake. There’s no easy way in and even when you’ve spent time figuring out how FOAF works there’s no instant return. Plus, the chance of actually finding friends of friends gets slimmer with each connection because each of your friends must be similarly interested in playing with XML for the network to expand. This difficulty is a crucial hurdle for a social network tool. The audience of people who could understand FOAF in its current state, and have the time and energy to work it out, is orders of magnitude smaller than Friendster’s potential market.

Given that FOAF is open and grassroots and therefore Good, what’s the solution? Two immediate ideas spring to mind. Frienster is currently very closed, and as Ben Hammersley says there’s no way to get at its data. While this is probably a commercial decision, designed to protect the data the company is accumulating, it could be a mistake. Opening things up, letting people play with the data, could ultimately attract more users while a proprietary system could wither due to a lack of innovation. The simplest route would be for Friendster to automatically generate a FOAF file for every user. The FOAF network would immediately explode in size thanks to a few hours of coding. Add the facility to host FOAF files elsewhere (for those that want to), or import existing files and Friendster is suddenly the default interface for FOAF. (Some, however, might think this commercial co-opting of a fledgling open technology could be too much too soon for FOAF.)

The other solution is for someone to create an open Friendster-like front end for FOAF from scratch. Obviously, this is more work and, given the nature of geeks, it could end up being far less usable than Friendster. This is, however, what FOAF really needs if it’s to take off. If such a solution appeared now it might be the wrong time; after Friendster the audience may be social-networked out and take-up on an open alternative might be slow. But this could be far more innovative and exciting than a closed commercial solution features that may see it through in the long term.

A social network tool needs to be usable by everyone in the network or else its usefulness diminishes rapidly. Currently, Friendster achieves this but lacks in openness and potential innovation. FOAF somehow needs to reach this level of usability at which point we can hope its other benefits will become a winning advantage.