facebook data store api thoughts
On Monday, Facebook employee Haiping Zhao quietly published a Data Store API on the Facebook developer wiki. — It's preliminary, and still in closed beta, but it marks a milestone in the evolution of the Facebook platform.
Context & Ripple Effects
Three months after the Facebook Platform launched — the rollout Marc Andreessen dissected three weeks in back in June 2007 — the stack is gaining its storage layer. Facebook engineer Haiping Zhao has posted a preliminary Data Store API to the developer wiki, still in closed beta, giving third-party apps a place to persist data inside Facebook rather than on their own servers.
The timing matters: the Application Directory already lists dozens of apps per category, analysts are framing News Feed placement as the next SEO battleground, and the company is fending off whispered multi-billion-dollar acquisition interest while preparing for an IPO. A hosted database is how a social graph company becomes a full infrastructure vendor.
First-order effects
- Facebook application developers can now store queryable data on Facebook's infrastructure instead of provisioning their own databases, cutting the operational cost side of building on the platform.
Second-order effects
- Rival social networks under pressure to match the platform will need to offer equivalent hosting and storage APIs or watch developers concentrate where the whole stack — distribution via the feed plus persistence — lives in one place.
Third-order effects
- Once application data lives in the platform owner's store, the platform operator controls both the code and the data layer, making future decisions about who may access that data — and on what terms — a structural lever rather than a technical one.
The trend: Social platforms are extending from application containers into full-stack infrastructure providers, with the platform owner becoming gatekeeper of storage as well as distribution.