Loosing Site

The web could become a friendlier place thanks to Google’s Social Graph API. This new service helps developers build applications that query public information about people and their relationships, saving users the effort of redefining their connections as they move from community to community, and from application to application.

Developer Home – Google Code via kwout

This information is gathered from public profiles on existing social networks, from blogs and from suitably formatted web pages. The systems understands and indexes data coded in XHTML Friends Network microformat

XFN: Introduction and Examples via kwout

or marked up with Friend of a Friend syntax

FOAF Vocabulary Specification via kwout

to define relationships between people, or to define “me” relationships between an individual’s multiple profiles held on different systems.

In the video below Brad Fitzpatrick describes the API and some of its use cases.

Building on a growing base of semantic web technologies this service is another clear pointer to the direction the Internet is taking, with markup focussed on meaning rather than presentation, and applications able to work with and augment each others data as a matter of course. The API will affect social networks, who will need to choose the level of openness they are willing to support.

Though the system uses only public data, the power to query and analyze could affect the way you choose to present yourself online. It could also lead to some interesting situations where connected people perceive different qualities in their relationships. It’s not just how you link to your friends, it’s how they link to you. Are your ears burning?

Reuters have launched OpenCalais a service that will encourage the development and growth of the Semantic Web.

OpenCalais via kwout

Aimed at developers and information publishers, OpenCalais is web service that can extract meaningful terms from well written text. When fed a newspaper article or business report the service will identify the people, places, monetary values, companies and institutions mentioned and will organize references to those terms to ease further research. In well structured business text Calais can also interpret and classify facts; recognizing and reporting joint ventures, mergers, acquisitions, earnings announcements etc.

End users will not interact with Calais directly, but will access the service through third party applications, plugins and utilities. Development of these will be encouraged through programming bounties, the first offers $5000 for the development of a Wordpress plugin.

In the long term, initiatives like OpenCalais will help us work more effectively with the increasing flow of information that surrounds us. These are distinct strategic and personal benefits to be gained from generating, and being able to work with, structured information instead of being swamped by raw data.