1 Proposal to update the code charts for Mongolian

The Mongolian encoding is complex and the code chart (UTC 2013b) does not adequately convey how users should access basic glyphs that are available via this encoding. One complicating factor is that, like Arabic, Mongolian script has positional forms for most letters. A second factor is that Mongolian words are subject to vowel harmony which means that within a single word, vowels and some consonants are consistently either masculine or feminine.1 The masculine and feminine letters can have different glyph shapes. There can also be other types of glyph variation. The Mongolian encoding relies on context and OpenType lookups to determine and select the correct glyph shape from the available options. Positional variants can be reliably invoked based on the position of a glyph in a word. For illustrative purposes, the positional forms can also be triggered using a Zero Width Joiner (U+200D) on one or both sides of a Mongolian code point. The masculine and feminine glyph variants cannot always be determined based on context. Therefore, three Mongolian Free Variation Selectors (U+180B, U+180C, U+180D) are included in the Mongolian block to select particular glyph forms. These variation selectors may invoke a specialized form of Mongolian letter. The combination of positional forms and glyph variants leads to a matrix of sixteen potential forms for each code point. There are no Mongolian letters that require all sixteen forms, most need significantly fewer. One of the most complex letters is U+182D MONGOLIAN LETTER GA