Lesson 15

JSON in Today's Ecosystem

Strengths, limits, and when another format might fit better.

JSON won mindshare because it is simple, text-based, and maps cleanly to object models in mainstream languages. It is not the best choice for every problem—understanding tradeoffs helps you pick formats deliberately.

Where JSON fits well

  • HTTP APIs with JavaScript, mobile, and server clients
  • Config that tools parse at startup (tsconfig, CI matrices)
  • Event streams when each message is a self-contained record
  • Interoperability when you cannot ship a shared binary schema to all parties

Friction points

LimitationPractical impact
No commentsDocument fields elsewhere or use JSONC locally only
No dates or decimals as native typesEncode as strings with agreed formats
Verbose vs binaryHigher bandwidth than Protobuf or MessagePack
Schema optionalDrift between producers and consumers

None of these disqualify JSON—they define where extra discipline (schemas, tests, docs) is required.

Neighbors in the format landscape

  • XML — Still strong in document-centric and legacy enterprise systems
  • YAML — Human-authored configs; watch indentation and security on untrusted input
  • CSV/TSV — Flat tabular data, not nested graphs
  • Protobuf / Avro — Compact binary with strict schemas inside trusted networks

Teams often use JSON at the edge (public API) and binary formats internally.

Adopting JSON thoughtfully

Standardize on UTF-8, publish schemas or OpenAPI where possible, and version breaking changes explicitly. JSON’s ubiquity is a social convention as much as a technical one—your course knowledge lets you participate in that convention without treating the format as magic.

When you want to practice, use the related DevCove tool — optional, not part of this lesson.

Open related tool

Back to course overview