Christopher Herot's Weblog

OpenSocial Critiques

Dare Obasanjo has written what is probably the most detailed critique of OpenSocial that I have seen so far.  Most of his post enumerates the things that are still missing from the documentation, such as the Activities Data API, the Service Provider Interface (SPI), or any details of how the user can control the applications access to personal data.  He also brings up some more substantive points:

• No user interface hooks, even for common operations such as interacting with a Friends list
• The data in represented as Atom feeds, which can be very convoluted.
• The persistence data API only supports key-value pairs.
• The spec is not "open" in the sense that it is owned by a standards body instead of Google.
• There is no mechanism to map identities among different container sites.

Of course this last point does create an opportunity for application developers.

We all await developments on the Google OpenSocial blog.

Other opinions:

• Danny Ayers
• Russell Beattie
• Shelly Powers (also here)

Also, see Niall Kennedy's survey of widget formats.

More on OpenSocial

Kevin Cawley points out a few potential flies in the ointment of the soon-to-be-announced OpenSocial API.  In a particularly apt analogy he reminds us of the early promise of Java:  write once and run anywhere.  The reality was that while Java turned out to be a perfectly fine programming language, there was no way to abstract out all the differences among platforms while still taking maximum advantage of the unique capabilities of each.  The same thing is likely to happen as developers attempt to adapt their applications to the multiple social networks that will be behind the OpenSocial API.

Cawley's prediction is that Orkut will enjoy a momentary increase in usage as developers use it as a sandbox but then those same developers will get bogged down in trying to expand their app to the other platforms.  I am less concerned about that - any developers who sanely wanted to diversify across multiple platforms would have to write their own abstraction layer if they didn't use OpenSocial - it will be interesting to see how closely these competing SNS companies will adhere to the emerging standard.

OpenSocial details leaking out

The New York Times and TechCrunch both reported more details today on the OpenSocial set of APIs that will unify All Social Networks (except Facebook and MySpace).  What's especially interesting is the list of SNSs that have signed on.   In addition to the usual suspects (Orkut, LinkedIn, hi5, Friendster, Plaxo, Ning) the alliance includes Oracle and Salesforce.com.  While as a developer I look forward to leveraging all these sites through one convenient API I do wonder how Open Social will deal with the varying policies of the member sites.  For instance, LinkedIn has said they will require applications and/or developers to be certified before admitting them to the system.  How will these differences, and the different underlying features be exposed to the end users?