How Facebook Ships Code
I'm fascinated by the way Facebook operates. It's a very unique environment, not easily replicated (nor would their system work for all companies, even if they tried). These are notes gathered from talking with many friends at Facebook about how the company develops and releases software.
Context & Ripple Effects
This piece lands at a charged moment for Facebook: days earlier, Goldman Sachs clients bought $1.5B of stock at a $50 billion valuation and Wedbush's Lou Kerner argued a public market would price it nearer $100B, so outside observers are dissecting every layer of how the company actually works. FrameThink's notes on the development process extend an existing line of coverage — Royal Pingdom mapped the software behind the world's largest site in June 2010, and GigaOM covered how Facebook squeezes more from its machines in late 2009.
First-order effects
- Facebook engineers can move a feature from internal commit to full-user rollout quickly, with the Prototypes program letting staff float experimental features to opted-in power users before committing them site-wide.
Second-order effects
- Rivals like Google, already competing with Facebook for ad dollars against eMarketer's estimated $1.86B in largely small-business revenue, face pressure to match a release cadence that treats shipping speed as a product feature.
Third-order effects
- Publishing the process itself — as FrameThink does here — turns engineering culture into a recruiting and brand asset, pushing continuous-deployment practices from startup curiosity toward an industry template other web firms are expected to answer.
The trend: Leading web companies are treating their internal development and release process not as back-office plumbing but as a core competitive advantage worth documenting — and imitating.