![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Google's OpenSocial is more widget spin than real, usable cross-social network API.
I rip it a new one over at IT Pro:
I rip it a new one over at IT Pro:
I've been looking at Google's OpenSocial, and to be honest, I'm not particularly impressed. There is a need for a way of bringing the spiralling maelstrom of social networks into some coherent, cohesive whole. I can see how easy it is for people staring at the headlights of the oncoming Facebook juggernaut to want to seize hold of the first possible escape route - it's just a pity that OpenSocial came along first.Read more at IT Pro.
Yes, OpenSocial can be used to extract information from different web sites and bring them together, but it's missing many of the features that would make it truly compelling.
Firstly, and critically, there's no identity component to OpenSocial. All it is is a set of simple API calls that extract all the information that's available. There's no way for a data provider to control just who sees what and how,it's an all or nothing system. The simplistic model that OpenSocial currently offers means there's no way for me to set a set of rules that expose information in different ways for different people, which is something that's critical when sharing information across sites - which is something that I see as vitally important, and I'm someone living what can best be described as a "radically transparent" online life.
busy
(no subject)
And you have a good point about what we need - but I wasn't expecting OpenSocial to provide those things. For steps in the right direction of that take a look at:
http://www.sixapart.com/about/news/2007/09/were_opening_th.html
and
http://www.sixapart.com/about/news/2007/10/a_river_of_rela.html
(no subject)
LJ makes a stab at dealing with it with groups, as will Facebook soon, but that's not really enough.
My gut feeling is that, as nice as the idea of bringing together social networks together is, the web is not going to provide the solution. Instead what we'll end up with is something on the desktop that mines the information we have locally and on line, and uses a rules engine to build our own personal social application, based on what we want to know, when we want to know it.
(Of course that raises the whole hoary chestnut of context computing, but that's a drum I've been beating for the last 10 years)
(no subject)
(no subject)
The API looked rather disappointing, too.