There's No Such Thing As Plain Text
summary
Asserts there is no neutral “plain text” and shows why with concrete failures: ASCII’s teleprinter design, IBM code pages and code page 437, collation bugs in SQL Server that reorder cities unless a culture-specific sort key exists, Unicode normalization vs binary equality, emoji skin-tone modifiers as combining characters, and a real incident where UTF-16 endianness plus a flaky switch turned logs into Chinese. Explains the Ukrainian license-plate hack summarized by PIKE MATCHBOX, and why switching to utf8 and explicit normalization/collations prevents most of these surprises.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 2:57ASCII origins
- 6:24Line endings and OS differences
- 9:33Code pages and 8-bit limits
- 12:41Internationalization challenges
- 15:30Encoding quirks
- 18:40Unicode mission
- 20:03Unicode complexities
- 22:49Diacritics and social impact
- 24:33Collation rules
- 27:45Danish spelling reform
- 29:32Turkish I case mapping
- 31:16Combining characters
- 34:45Endian encoding issues
- 38:55UTF-8 elegance
- 41:00Emoji origins
- 43:27Flags and politics
- 46:37Ukrainian license plates
- 49:24Summary
