SRE's organizational position shapes how it communicates. It is **not a command-and-control organization** — an SRE team owes allegiance to **two masters**: the product development team whose service or infrastructure it supports (a strong relationship, because SRE is held accountable for those systems' performance), and the SRE organization as a whole (the actual reporting line, source of shared culture and homogeneous approaches). This dual allegiance makes communication and collaboration first-class concerns. The book's governing metaphor: **a team's interface to other teams is like an API.** Data — about projects, service state, production, and individuals — must flow reliably from one party to another, just as data flows around production. And like an API, **a good design is crucial and painful to correct later**, so the interface a team presents deserves deliberate design. The same API-as-contract framing applies to collaboration. SRE's distinctive contribution is the *mix* — software engineering skills, systems engineering expertise, and the wisdom of production experience. The promise: an org charged with reliability that has the same skills as the product teams will improve things measurably. Merely putting someone "in charge of reliability" without the full skill set is not enough.