Google: Don't give up on OpenSocial

When Google unveiled its OpenSocial developer initiative at the end of October, observers hailed it as the future of the social Web.

But is the search king already too late to the party?

It's been over six weeks, and OpenSocial--which uses open-source code to allow any participating social media site to implement a common set of application program interfaces (APIs) and create "universal" applications--isn't finished, though developers believe it will be ready early in 2008. In the meantime, a number of partners have launched independent developer platform strategies, and Facebook has announced that other social networks will be able to use its own applications, rivaling what Google can offer.

Earlier this week, three OpenSocial partners--LinkedIn, Friendster, and Bebo--made their own platform announcements, and the only thing they could say for sure about OpenSocial was that they'd start implementing it as soon as it was ready and stable.

Now some critics are saying if Google's OpenSocial group doesn't move quickly, its work could get lost in the clutter.

"There's a riff that OpenSocial could die on the vine," said Forrester Research senior analyst Jeremiah Owyang. "There are already multiple APIs, and it really is going to come down to what are developers going to choose to use, which one is going to be easier for them."

Not surprisingly, OpenSocial developers insist that the initiative isn't lying fallow. "We've made a lot of progress," Joe Kraus, director of product management at Google, said in an interview with CNET News.com. "In the classic kind of platform world, what typically happens is there's a lot of work that goes on that isn't consumer-visible."

Social-networking sites have been scrambling to develop "platform strategies" in order to compete with Facebook, which first invited third-party developers to create custom applications for the site in May. Facebook now reports that over 100,000 developers have contributed to the service.

OpenSocial was the only potential rival to the Facebook platform. For many smaller social networks, as well as larger ones grappling with the technology to build developer platforms of their own, OpenSocial was (and certainly could still be) a ticket to success. Not only did it have the Google influence, it also came with just about every social network not named Facebook, including big names like MySpace, LinkedIn, Plaxo, and Friendster.

Some OpenSocial partners say they're happy Google announced what it was working on before completion. "It's a good thing it wasn't announced any later," said Joseph Smarr, chief platform architect at Plaxo, which has been an ardent supporter of Google's platform. "Facebook launched their platform in May, so for the past six months before OpenSocial launched, companies were thinking 'How do I respond to that?' and so they each came along with their home-brewed thing."

Kraus adds that some of the independent platform strategies would be necessary even if OpenSocial were finalized. One of them is LinkedIn's "InApps," which also aims to spread LinkedIn's data and influence outside the business-oriented social network through partnerships with other Web sites. "OpenSocial so far is really about how developers embed their application into a social network," Kraus explained. "A good chunk of LinkedIn's APIs is about how LinkedIn extends their social-networking data into other sites."

But this week, Facebook stirred the waters even more. The real twist for OpenSocial came with Bebo's announcement on Wednesday: The youth-oriented social network is launching an independent platform prior to OpenSocial's debut and taking advantage of Facebook's own APIs to make it easy for a Facebook developer to convert its applications to Bebo applications.

Bebo co-founder and CEO Michael Birch said developers from the social network had approached Facebook to see about application portability, and Facebook was receptive. "A lot of our developers know developers at Facebook," he said in an interview. "We kind of opened up to them and they were very warmly receptive to the idea."

Application manufacturers were giddy about the possibilities--casual gaming site Bunchball, for one, reported that with Bebo's implementation of Facebook's APIs, Facebook members would be able to go head-to-head with Bebo members in its games, an early sign that perhaps true interoperability across social networks could be on the way.

For Facebook, Bebo is just the beginning. The company quietly announced later on Wednesday that its developer platform would be further opened for use on other social networks, in effect creating an OpenSocial of its own. Facebook is no Google, but its platform has several months' head start and a markup language that many developers are already well-versed in.

Still, if Facebook has a growing track record for social applications, Google has a track record for just about everything else. "We don't have any philosophical objections" to Facebook's APIs, said Plaxo's Smarr, but OpenSocial may be a better choice because "it's all built on existing open standards like HTML and JavaScript." Google may provide better resources, too. "Google's doing a whole lot of work to provide open-source code for developers and container implementers. They're really putting their muscle out there to help people," he said.

One thing's inevitable: complication. Even a single universal standard won't be a complete cure for a social network's platform woes. "Applications will never work the same over any community," Owyang acknowledged. "The reason is, every community has different technographics (the hardware, software, and code used) and different demographics. They use the tools differently. To expect that one widget will work cleanly over all platforms is overoptimistic."