It's 2026, and most mobile apps still serialize data the way they did in 2001: JSON — a format borrowed wholesale from JavaScript, by developers who often can't stand JavaScript. This talk takes a quick tour through serialization: what it actually is, a bit of history on how JSON became mobile's default, and why it won in the first place — human-readable, ubiquitous, zero friction with web APIs. Then we get honest about what it costs us: verbose payloads, slower parsing, no native type safety, and real battery and bandwidth impact at scale. From there, we look at the alternatives, landing on Protocol Buffers as the main event. Using simple Android/Kotlin snippets, we'll compare everyday JSON handling (kotlinx.serialization, Moshi) against protobuf side by side, and see what schema-first, binary serialization actually buys you in a real codebase. One question to leave with: are we still using JSON because it's good, or because it's just what we've always done?