- 2, “gujarâtî”: Use a consistent transliteration scheme throughout the document. - 3.4.4: The spelling alternation is not relevant. Both functions are representation of a nasal sound. - 5.2: Why are U+0A8C GUJARATI LETTER VOCALIC L and U+0AC4 GUJARATI VOWEL SIGN VOCALIC RR included? Don’t they belong to the same category of excluded letters vocalic rr and vocalic ll? - 5.5: It’s actually just as simple as: `C[N][M][B|X] | V[B|X] | C[N]H` (consonant clusters can be broken down to multiple preceding occurences of `C[N]H`, when the exactly rendering of a cluster is not the discussion’s concern. - 6, “There are no characters/character sequences in Gujarati, which can be created by using the characters permitted as per the [MSR] and look exactly alike.”: Should be MSR and WLE (which restricts the cluster structure, preventing sequences like `VM`). Best, 梁海 Liang Hai https://lianghai.github.io