What I want instead of Facebook’s OpenGraph

Facebook makes a great pitch, don’t they? One line of javascript. More user engagement. Viral distribution on the world’s largest social network. It’s compelling.

The downsides, though, are worth examining. Limited access to your visitors’ contact information means that converting visitors into repeat customers is not necessarily easy. Not all of your visitors use Facebook, and might want to connect profiles from other social networks or public profiles. Further, with Facebook positioned in the center of the exchange, what happens if Facebook suffers access problems? Facebook could become one of the largest single points of failure on the internet.

While their offering from this week’s F8 event are compelling – and I’ll likely add some of their tools to my sites – I want something else. A set of easily plugged-in tools not tied to any particular social net, but which can access any or all social nets – not unlike the OpenLike project, but bigger.

For example, Facebook’s social plugins allow my site’s visitors to see what their FB friends are doing on my site and discover more of my content which might be relevant to them, based on their social graph. Let’s consider what would happen if the plugin was network-neutral, though.

Visitor arrives for the first time and has the option of selecting one or more social profiles to connect with my site. The visitor is in charge here – they can decide how much of their personal life (which networks) to share with my site based on their own privacy concerns and on how much value they want to get from my site. This is a win for the visitor, who gets to add value to their experience on my site without compromising privacy or having to sign up for yet another account, and the publisher, who gets more engagement on her site and access to a diverse range of social networks on which to distribute her content.

Once the visitor selects one or more profiles, the plugins go to work aggregating information from the selected nets using their APIs. It would be preferable if the nets agreed on a standard, open set of APIs for social graph access.

One piece of this semi-federated system is that the publisher needs to be able to store some of the visitor’s information. Again, this would have to respect their privacy settings, but publishers need the ability to convert visitors into customers and for most of us, that means email. If a visitor agrees to share their email address with my site when they connect their profile, I should be able to retain that information and offer them subscriptions to my email services (as a start).

Jeff Jarvis takes a slightly different direction, from the POV of the visitor/user, in his discussion of the Bizarro Facebook. In Jeff’s idea, visitors can tell the social nets where to go to learn about Jeff.

It’s entirely possible that what I’ve described already exists. RPX seems to have an API capable of doing more than just allowing logins to a site from a diverse group of social nets. Also, what about OpenSocial? Wasn’t it supposed to offer much of this?

The bottom line is that publishers probably aren’t going to get a great deal from Facebook and aren’t going to be thrilled in handing over so much of the customer relationship to Facebook. Many (most?) will sign on and use the new Like button, but how many will opt for deeper functionality? And how many would if it meant they weren’t locked in to a single social net partner?