Against fancy ligatures in programming fonts
3 months ago
- #typography
- #programming
- #unicode
- Ligatures in programming fonts are criticized for potentially misrepresenting code semantics or causing reader confusion.
- Ligatures combine troublesome characters into single glyphs, but in programming, they can introduce ambiguity with Unicode symbols.
- Programming ligatures are 'dumb' substitutions that ignore semantic context, leading to errors in code interpretation.
- Unlike in body text, where ligatures like 'fi' have clear meanings, programming ligatures can be semantically wrong depending on context.
- The author argues against ligatures in programming fonts, emphasizing the importance of character disambiguation for code clarity.
- Examples from Fira Code illustrate how ligatures can confuse existing Unicode symbols, creating unnecessary ambiguity.
- The article mentions LaTeX users incorrectly substituting curly quotes for backticks, highlighting similar issues with semantic accuracy.
- The author concludes that for code meant to be read by others, avoiding ligatures is crucial to prevent unnoticed errors.