Tony Karrer's eLearning Blog on e-Learning Trends eLearning 2.0 Personal Learning Informal Learning eLearning Design Authoring Tools Rapid e-Learning Tools Blended e-Learning e-Learning Tools Learning Management Systems (LMS) e-Learning ROI and Metrics

Wednesday, October 31, 2007

OpenSocial Platform

Well, we finally are starting to hear what Google's anticipated response to the fact that Facebook Platform and Facebook as a Learning Platform. The real beauty of Google's approach is in the word "open." With Facebook you are somewhat forced to make a hard choice about rewiring your application to live seamlessly within the Facebook environment or pushing Facebook users across to your site which is a bit ugly. Further, you face the prospect of having Facebook rework aspects that you are leveraging.

So, as I mentioned before, I'm a big believer in leveraging the knowledge of social connections that exist across the network of different applications (e.g.,, flickr, MyBlogLog, Facebook, LinkedIn, your email, discussion groups, etc.). Google appears to be headed down the direction that will allow applications to take advantage of that knowledge.

But I'm still back at my same question:

I'm currently in process on two Facebook applications that leverage that platform for viral expansion via Facebook, and I'm working on the design of several new or extensions of existing sites/software solutions that will leverage social aspects. The questions are:
  • Do you leverage Facebook because of installed based and easier, known viral adoption model even though you are locked in?
  • Can you justify building out your own social graph (friends, groups, etc.)?
  • Do you head towards Google's OpenSocial even though you may be ahead of your users?
Answers seem to be fuzzy and turning out different in different cases. I'm hoping to hear more from other folks about what they seem happening.


Anonymous said...

An interesting and relevant question, I think. One area that is offering room for improvement right now, which may appeal to you, is the "unified graph" idea. Namely, figure out some way to let users import their entire graph into one place that users and developers could then use and reuse however and wherever they like.

It seems to me that if your users could take their friends out of Facebook, any other online services they use (email, OpenSocial apps, etc), easily manage that, then give your applications access to only what you need, that solves a lot of your questions. No more lock-in at Facebook, reduced fatigue on your users in finding and inviting their friends.

This happens to be something I'm thinking about a lot lately, because I'm helping to build a web service that would do just that.

From there it seems to me that the real question comes down to how to implement this kind of tool in a way that is easy and transparent for users and application developers. Ultimately it all boils down to serving up an Atom or FOAF document to an application with data the user wants to share.

My team and I have a handful of interesting ideas we're pursuing but we'd love some feedback — what would make sense to an application developer such as yourself?

Tony Karrer said...


I think having the ability to access the social graph easily is the key. However, the problems of uniquely identifying across platforms, controlling access, etc. seem a bit hard right now. So, I'm not sure I quite get your question. You can serve up people, friends, groups fairly easily.

The applications we are looking at rely on accurate access to that data to then be able to build social applications on top. That way you don't have to deal with re-establishing your social relationships.

That said, you would need to have fairly flexible ways for people to specify which of their friends they want to bring across.