Tag Archive for 'opensocial'

The difference between a widget and an application

The previous post got me thinking of another outcome due to the differing functionalities of the two platforms. The concept of what a widget is and what an application is becomes clearer:

  1. 2nd generation widgety microsites in widgety-sized living spaces
  2. engaging and useful applications which make use of your digital web of friends.

    This second type of application can be split into another couple of sub categories:

  1. Utility apps – Training regimes (Nike+) and or niche digital organisers, a bit like ‘calender’ on Steriods which benefit aspects of your real life.
  2. Games – Engaging multiplayer, multi-platform, geo-positioned, haptics-enabled bundles of fun to be shared with friends and family. Where you can put the age old disciplines of human competitiveness, cooperation and comparison to the test. And before you ask, yes, “Compare”, “Crush” and “Growing Gifts” are all games.

The most successful of these will create an experience which offers both usefulness and enjoyment. Nike+ is sure to fare well, but who says there won’t be other Nike applications in the future? How difficult would it be to imagine an entire suite of specialist applications all created by Nike and sitting on your desktop…

As Nokia demonstrated at the Games Developers Conference this week, along with rumours of a partnership between Apple and games company Gameloft, we shouldn’t think that the internet is the only place, or the best suited, to socially interact.

facebook vs. opensocial

Shuzak’s presentation is very interesting.

One slide particularly got me thinking about some of Facebook’s recent moves. In the last couple of weeks Facebook has begun to put the kosh on invites in order to create more thoughful and considered applications and drive the spammers away.

I thought OpenSocial’s achilles heel was this lack of ‘virality’. It’s possible to create generic applications, but OpenSocial cannot take advantage of Facebook’s social graph. It is however scalable. And that’s the irony, because if you build something on Facebook that works through the social graph, it will be impossible (at the moment) to get it working across OpenSocial – All those viral invitations useless…

By pushing back so heavily on the invites, Facebook is helping developers not to think about how to game viral goodness out of the platform, but instead how to build really engaging applications. It also let’s any particularly successful applications that work on OpenSocial to benefit from additional engagement with the social graph rather than being spammed with it.

Despite Facebook’s very clean and well documented platform, the numbers still matter, and although the experience and audience will be more engaging and valuable on Facebook, it’s dwarfed by the 250+ million accessible users available on OpenSocial. I’d hate to be an OpenSocial developer right now though…it looks painful!

Undoubtedly there’ll be experiences created for both platforms, but I doubt very much whether there’ll be one killer branded application that’ll work just as well on either…unless it’s a mobile one…who could forget Tetris :)

Google’s OpenSocial API

Google’s announced OpenSocial, an API which enables applications to be wrapped in a universal API which allows them to work across a number of different social networks. And all I can say is so what?

Google’s OpenSocial API is just a common Google-sponsored widget format for mini-applications. One which Google owns the control on. Sure, anyone can write an app using OpenSocial, but anyone can write an app on Facebook too. It looks just as proprietary as the Facebook system it more or less copies.

What’s crucially missing from this solution is openness – What OpenSocial offers is a way in which any application can be wrapped in a container API. That container API is tweaked and made to fit with each of the different social networking solutions out there. This means that the code in the app doesn’t need to change; if the app needs a list of my friends it just calls:


If my app sits on my Hi5 page it will get those friends one way and if I have the app on my MySpace page it’ll get my friends another way. The app itself doesn’t care how that happens – that’s the Container API’s job.

The Container API has not been mentioned, it’s the elephant in the room. With access to it I could create a social network on my blog…or Nike could sign up to it and ensure that already popular social apps could fit into their site furniture on their domain, rather than having to fit into another social network’s furniture e.g. Sponsored pages on MySpace. Powerful eh? You can appreciate why it hasn’t been mentioned given the speed at which they’ve released OpenSocial this is clearly an aspect of the API which needs a little more thought and consensus.

It’s the 1st question to ask when considering a brand such as Nike and whether they’d want to sign up to OpenSocial in the future, and that’s the most exciting thing about Google’s announcement of OpenSocial – It stops us from chasing functionality and concentrating on the basics, WHATEVER the social network happens to be.

The conversation is finally moving towards the root of social networking, and I’m all for that.

Get Adobe Flash player