The summer of Facebook is turning into the fall (as in the season, not the descent into an abyss) of social networking platforms. Facebook and its CEO Mark Zuckerberg are still the prince of social networks. The Graphing Social Patterns conference earlier this month explored multiple aspects of the FB and social networking phenomenon. Coming up this week is SNAP (Social Network Application Platform), a one-day conference focusing on the Facebook platform and beyond.
At the Web 2.0 Summit last week, Zuckerberg was a star attraction, but didn't have any surprises, such as Microsoft acquiring a large, expensive chunk of his company. About the only noteworthy nuggets from his time on stage were that Facebook would more than double its staff in the next year and have more to say about its targeted ad system in the next few months.
The big news at the Web 2.0 Summit came from MySpace, which is following in Facebook's footsteps and opening up its platform to developers. MySpace's platform will have a set of APIs and markup language for developing applications that run within the MySpace environment. Like Facebook, MySpace plans to lets users categorize friends as family, business, etc. In addition, the company is also developing a catalog for MySpace widgets in the next few weeks.
Google is set to announce APIs on November 5 that will allow developers to tap into the Orkut social graph and iGoogle.
Linkedin announced that it will give developers some access to its platform, adding new capabilities to Linkedin and allowing members to tap into their connections from services such as Salesforce.com.
The trend is clear and emergent. Building platforms, not just applications or services, that attract developers is the way to succeed. The more developers, the more a platform potentially has to offer its users. Salesforce.com, for example, is betting its business on this concept. Microsoft has been doing it for years. Sun with Java the same. Every open source project. SAP with NetWeaver. Even the famously closed Apple is going to open its iPhone to developers.
Image by Alden S. Klovdahl of Australia National University
However, we end up with a multitude of incompatible platforms. Developers either have to create separate applications and widgets for each platform or pick only a few to support.
Dare Obasanjo writes that Facebook is the hands down leading social networking platform for good reason and that those who follow need to meet the following criteria: the ability to monetize applications; distribution and reach to stimulate viral growth; access to users data (the social graph); the ability to build an integrated and immersive experience; and shielding smaller developers with successful applications from cascading infrastructure costs (the Winner's Curse).
Adoption of some cross social network standards would also be an important criteria for reducing development friction. “We want to get to a point that just as blogs sites can’t launch without RSS or Atom, a social network can’t launch without an analogous data interchange format,” said Brad Fitzpatrick at the Web 2.0 Summit. Along with David Recordon, Fitzpatrick developed the Six Apart Relationship Update Stream, a feed of social relationship data that allows Web services to send and receive data when changes to social relationships occur on their service. Magnolia and LiveJournal support it now, and Hi5, Plaxo and Six Apart’s Vox will add support soon.
Joshua Allen, a Microsoft evangelist summarizes the current situation that developer face:
To be fair, I do believe that "the web is THE platform", but I think Google is being a bit disingenuous by trotting out the aphorism in this context. The web is the platform upon which many different and incompatible data gardens will do battle. There WILL be fights, winners and losers, and this will happen BECAUSE the web is the platform. If anyone is really serious about reducing the bloodshed and adhering to the spirit of the web, they should put their money where their mouths are and let loose some of the control from their own walled gardens and incompatible schemas.