ESSENTIALS - There's No Such Thing As Plain Text
summary
Claim: there is no such thing as plain text. Traces from 1960s ASCII and IBM code pages cause real failures: CP437 box‑drawing, passport names changing Å→AA break borders, Danish collation needs extra sort columns. Encoding bugs include UTF‑16LE mistaken for UTF‑16BE and a faulty switch that drops a byte and turns logs into Chinese. Unicode normalization (NFC/NFD), UTF‑8 adoption, ZWJ skin‑tone modifiers for emoji, and the license‑plate trick PIKE MATCHBOX are the concrete tools and gotchas shown.
AI-written summary · about
chapters (21)
- 0:00Introduction
- 1:28Myth of plain text
- 3:12ASCII origins
- 6:23Carriage returns vs line feeds
- 8:53Code pages failure
- 13:06Text encoding mismatches
- 16:47Unicode birth
- 18:51Hungarian alphabet
- 21:17Diacritics and ligatures
- 23:42Passport name mismatches
- 24:44Collation across languages
- 28:12Politics in orthography
- 28:54Unicode normalization
- 32:05Debugging Unicode error
- 33:30UTF-16 endianness
- 35:13UTF-8 backward compatibility
- 37:19Emoji as alphabet
- 39:27Flags and encoding
- 41:55Ukrainian license plates
- 43:40Pike Matchbox reality
- 44:01Conclusion
