CharBusters - 10 Unicode Myths
summary
Unicode hides gotchas: CJK already needs ~88,000 code points, Java char and String length are misleading — String.length() measures code units (surrogates), not user characters, Java 9 switches to compact byte[] storage and getBytes() uses platform default so encoding can silently change, use explicit UTF‑8. Emoji and flags are multi-codepoint graphemes (Poland flag = two code points = 64 bits; some family emojis use 7 code points). Normalization (NF_KC) and grapheme algorithms fix comparisons; unhandled cases have crashed phones and even caused fatal miscommunication.
AI-written summary · about
chapters (13)
- 0:00Introduction to Unicode myths
- 2:24Myth 0: Unicode is solved
- 7:54History of ASCII and Unicode
- 8:36Myth 1: One character is one byte
- 13:58Myth 2: One char is one character
- 23:40Myth 3: Java is UTF-16
- 26:06Myth 4: Unicode is unambiguous
- 28:56Myth 5: Emoji limitations
- 36:01Myth 6: Length is useful
- 41:00Myth 7: Whitespace is straightforward
- 44:30Myth 8: Uppercase is simple
- 45:57Myth 9: Unicode is harmless
- 49:48Final thoughts and testing
