L2/04-418 November 18, 2004 Report of the Indic ad-hoc TDIL has asked repeatedly for a number of characters to be encoded, as well as various other modifications to the encoding of the Indic scripts. Document L2/04-398 is the latest iteration of those requests. However, the TDIL proposals are not presented in a form which can be acted upon, given knowlegde of the UTC outside TDIL. The members of the indic@unicode.org list have identified the individual proposals (see ), but so far did not succeed in formulating appropriate proposals. We should consider using our PRI process to progress the issues which seem to be the more mature and the most pressing. The rest of this report is proposed text for PRIs. 1. PRI: script specific Indic dandas The danda and double danda characters are used in a number of writing systems of South Asia. Currently, Unicode considers that U+0964 DEVANAGARI DANDA and U+0965 DEVANAGARI DOUBLE DANDA are not specific to Devanagari and can be used with other scripts. This is documented in TUS 4.0 (end of sections 9.2, 9.3, 9.4, etc). These shared characters may cause difficulties when handling multi-script texts, where it may difficult to determine the most appropriate font to use to render an occurrence of U+0964 or U+0965. The questions for reviewers are: should there be separate, script-specific coded characters for the dandas? For which scripts is there evidence of use of the dandas? Reviewers are encouraged to include evidence (scans of printed books, newspapers, etc.) in their feedback. Evidence in the context of multi-script texts is particularly relevant. 2. PRI: Eyelash ra The Marathi and Nepali languages, written in Devanagari, use the eyelash ra. This is currently reprepresented by the sequence . The sequence is also supported for compatibility with Unicode 2.0. The questions for reviewers are: should the eyelash ra be encoded as a separate character? Why? If so, how should the legacy issues be handled? 3. PRI: Bengali Yaphalaa The Bengali yaphalaa sign is currently represented in Unicode by the sequence . It has been proposed that a new character be encoded to represent yaphalla. The questions for reviewers are: should the yaphalaa sign be encoded as a separate character? Why? If so, how should the legacy issues be handled? [Note: we probably want to post Gautam's document as well, to give a basis for the discussion] 4. PRI: Malayalam chillus Unicode specifies that sequences of the form are displayed as chillus (for suitable consonants), that sequences of the form are displayed with a visible chandrakala on the full form of the consonant, but does not specify how sequences of the form (without any jointer) are displayed. The questions for reviewers are: what should the sequence without joiners render as? Alternatively, should the chillus be encoded as separate characters? Why? If so, how should the legacy issues be handled? 5. PRI: Gujarati fractions Characters have been proposed for encoding the 1/4, 1/2 and 3/4 fractions in Gujarati. The questions for reviewers are: should there be a single character, used in groups of one, two or three occurrences, or three separate characters? Should the character or characters be combining marks? Can you provide evidence of use? ----