Nepal Codes for Information Interchange White Paper v2 Font Standardisation Working Committee, 1997 page 10 varg 5, labial nasal consonant Ma non-varg consonant Ya non-varg consonant Ra non-varg consonant La non-varg consonant Va non-varg consonant SHa ¡ non-varg consonant SSa ¦ non-varg consonant Sa ª non-varg consonant Ha ® Conjunct ksha 4 Conjunct tra i Conjunct gya : glottal stop Ü Required for Limbu nukta àÕ Commonly used modifer for borrowed words anuswar à* Nazal modifier, may not be necessary, but part of the rendering. – equivalent to the nazal consonant of the varg of the consonant consonant modfiers CM1 CM2 CM3 CM4 For expansion of the codes to handle new languages Note that there are many letters in common between languages, but no language uses all the letters. It is this combined alphabet that will be encoded in the computer. There will be a Unicode table in which all vowels in their non-matra form follow their consonants and the implied vowel in the conventional writing systems is made explicit as the short “a”, $ . And there will be a “bridging” encoding for temporary use until support for Unicode becomes universally available. Simple translation paths should be available for conversion from existing TTF fonts to the bridging TTF encoding and also to Unicode when needed. 2. Sort orders. Let us use the Unicode encoding with the characters as in Table 3 and the notes following it. Then if we take the ordering of the individual letters of the alphabet as the sequence given, the sort order for words is obtained directly from the lexicographical ordering of strings over this alphabet. Let us look at a simple example in table 4, where the order is determined by the second letter of the internal form, where the sequence is $ ( Table 4 - An example of sorting. word romanised internal form sort order }ÍÉÈ (puriya) } ( & $È 2nd }Ée (pati) } $ e & 1st }Ãfz (prayatna) } $ $ e z $ 3rd