Blogs and geeks are buzzing (now you understand the picture above) about the release of FriendFeed’s web server code, named “Tornado“.
Tornado is an open source version of the scalable, non-blocking web server and tools that power FriendFeed. The FriendFeed application is written using a web framework that looks a bit like web.py or Google’s webapp, but with additional tools and optimizations to take advantage of the underlying non-blocking infrastructure.
Two of FriendFeed’s team members have moved their blogs to this new server technology.
The fact that anyone can download this piece of software opens the possibility that anyone can create the next FriendFeed, or maybe, if one stretches its developer’s mind, make it serve a thousand servers, talking to each other, finally giving way to the ever evading concept of Distributed Social Networking.
Of course, it’s not that simple.
Yesterday, Andy Oram posted a provocative piece on the O’Reilly Radar titled “RSS never blocks you or goes down: why social networks need to be decentralized“. In the piece, he writes:
Instead of the constant churning among the commercial sites du jour (Friendster, MySpace, Facebook, Twitter), the next generation of social networking increasingly appears to require a decentralized, peer-to-peer infrastructure.
Increasingly appears? I disagree. The appearance I get is that the big boys (Google, Facebook, Yahoo) have embraced a portability model where you can sign in and carry over some of your social graph to sites outside of their walled gardens. But the commercial sites are still very much in control and it increasingly appears it will stay that way.
These are some reasons why I believe commercial entities and not open source will still be the dominant gatekeepers of social networks:
- Regular people are not developers. The task of setting up a hosted web server, installing applications and maintaining them is not something that most people will be savvy enough to tackle. That leaves our desktops, but then there’s a lot of security issues running web servers from our home computers.
- Features are harder to implement if we have a protocol that everyone needs to adhere to. A commercial enterprise might want to push a new feature, work on it and release it, without asking the folks running the open source protocol to implement it.
- Money. There’s gonna be money involved in decentralizing social networks, and people are gonna have to pay it. And they won’t want to pay it. They’re already used to someone else (VC, Ads) paying the bandwidth costs for them.
- Trust. Whether we like it or not, people feel at ease once they know someone they can trust is running the show. If there’s no central figure in charge of the social interactions, users might not feel comfortable enough sharing their information.
Chris Messina talked about developers on a recent blog post:
If you wear the developer’s hat, now’s the time to get on board, read the specs, and implement support for OpenID, Activity Streams, OAuth, PubSubHubbub/rssCloud, or the other mentioned open standards that are relevant to your users.
Attacking that end should be no problem and the proof is on the pudding: brilliant developers like my friend @directeur are actively pursuing this route.
But trying to explain PubSubHubbub to my wife, and convincing her to leave Facebook — that’s going to be one-hundred times more challenging.
{ 7 comments }
