Some Thoughts on the Google Social Graph API
On Friday of last week, brad Fitzpatrick posted an entry on the Google code blog entitled URLs are People, Too where he wrote … I talked to Dewitt Clinton, Kevin Marks and Brad Fitzpatrick about this API at the O'Reilly Social Graph FOO Camp and I think it is very interesting.
Context & Ripple Effects
The arc here runs from August 2007, when Brad Fitzpatrick laid out the problem of fragmented identity in Thoughts on the Social Graph and reports surfaced that Google was working on a social network aggregator. After OpenSocial arrived in November 2007 as an application-level answer to Facebook's platform, the February 2008 launch covered by TechCrunch supplies the data layer: an API that indexes publicly declared relationships across the web rather than inside any one network.
Dare Obasanjo's take carries weight because he vetted the API directly with Fitzpatrick, Dewitt Clinton and Kevin Marks at the O'Reilly Social Graph FOO Camp. The pickup by Bokardo the same weekend shows the developer community treating this as more than a product note — it is a statement about where the social graph should live.
First-order effects
- Developers immediately gain a way to query public relationship markup (the XFN/FOAF-style links Fitzpatrick's post indexes), so any site that declares its connections becomes part of a queryable web-wide graph without joining a platform.
- Google positions itself as the neutral indexer of identity while Facebook's graph stays closed — the same open-versus-closed framing that drove the OpenSocial versus Facebook API debate in late 2007.
Second-order effects
- Closed social networks face pressure to either expose their relationship data or watch third-party apps route around them via the public graph, since the marginal cost of building on Google's index undercuts rebuilding friend lists per service.
- Sites and bloggers gain an incentive to mark up their pages with rel attributes and FOAF files, making structured relationship data an SEO-adjacent practice that Google alone controls the ranking and crawling of.
Third-order effects
- If adoption follows, the social graph shifts from a proprietary asset held by networks to web infrastructure indexed like pages — with whoever operates the crawler holding disproportionate power over identity, a role Google already occupies in search.
- The counterweight is behavioral: the API only sees relationships people publish in markup, so the structural question is whether meaningful social data migrates to the open web or stays behind login walls where crawlers cannot reach.
The trend: The social graph is being repositioned from a private asset owned by individual networks to public web data indexed centrally, with Google's crawler deciding how much of it is visible.