YAML, JSON & Markup Languages

Learn about modern data serialization formats (YAML and JSON) and markup languages (HTML and Ink). Understand when to use each format, how to parse them in different programming languages, and practice with real-world API data. This bridges structured data concepts with web technologies and interactive applications.

At-a-glance checklist

[ ] Understand YAML syntax: whitespace-based structure with key-value pairs and lists (dashes)
[ ] Know YAML use cases: config files, message passing, simple application saves
[ ] Recognize YAML limitations: ambiguous spec, tabs vs spaces, not as widely adopted as JSON
[ ] Master JSON structure: curly braces for objects, square brackets for lists, quotes for strings
[ ] Remember JSON quirks: trailing commas are INVALID, numbers don't use quotes
[ ] Know JSON is the standard for APIs: ~80% of web API responses use JSON
[ ] Parse JSON in multiple languages: Python (json.loads), JavaScript (JSON.parse), Java (verbose!)
[ ] Handle API pagination: APIs return metadata (count, next) + results array
[ ] Understand HTML basics: structure (<body>, <section>), presentation (, ), semantics (, )
[ ] Know the difference:  = italic (presentation) vs  = emphasis (semantic meaning)
[ ] Practice converting LaTeX to HTML (decide on heading levels)
[ ] Explore Ink for interactive fiction: knots (sections), diverts (links), choices (user decisions)

Key takeaways

YAML vs JSON: YAML is human-friendly (like Python indentation) but has ambiguous spec and tabs/spaces nightmare. JSON is machine-friendly, ubiquitous for APIs, and has clear syntax rules (but easy to miss trailing commas!).

Flexible schemas are double-edged: Both YAML and JSON allow different keys in different objects. Tom has favourite_numbers, Holly has favourite_food - that's valid! But your parsing code must handle missing keys gracefully.

JSON dominates APIs: If you're calling a web API, there's an 80% chance you'll get JSON back. It's language-agnostic (works in Python, JavaScript, Java, etc.) and easy to parse with built-in libraries.

API pagination is everywhere: The Star Wars API doesn't send all starships at once - it sends 36 results + a next URL. This prevents waiting "until the heat death of the universe" for massive datasets to download!