02 — Encodings

A String in memory has no encoding question to answer — it is UTF-16, always. The question arrives at the two boundaries: bytes coming in, bytes going out. Java's history at those boundaries is one long correction, finished in Java 18 and still visible in the API.

Lesson	The one thing
The default charset	UTF-8 since Java 18 — but native.encoding still follows your locale
Malformed input has two policies	the most obvious API is the one that hides corruption
The BOM is not stripped	three invisible bytes that break only your first line