Hangul normalization and collation is broken in Unicode, albeit for slightly different reasons. The Unicode Collation Algorithm explictly devotes two sections r

Hangul normalization and collation is broken in Unicode, albeit for slightly different reasons. The Unicode Collation Algorithm explictly devotes two sections related to Hangul; the first section, for "trailing weights" [1], is recommended for the detailed explanation. The Unicode Text Segmentation standard [2] explicitly mentions that Indic aksaras [3] require the tailoring to grapheme clusters. Depending on the view, you can also consider orthographic digraphs as examples (Dutch "ij" is sometimes considered a single character for example). [1] https://www.unicode.org/reports/tr10/#Trailing_Weights https://www.unicode.org/reports/tr10/#Trailing_Weights [2] https://unicode.org/reports/tr29/ https://unicode.org/reports/tr29/ [3] https://en.wikipedia.org/wiki/Aksara#Grammatical_tradition https://en.wikipedia.org/wiki/Aksara#Grammatical_tradition