L2/05-228 Subject: Analysis of a TDIL proposal for Vedic Source: Eric Muller Date: August 11, 2005 In early 2005, the Unicode Office received a large collection of materials from TDIL proposing the encoding of characters for Vedic Sanskrit text. This is the most recent version of series of proposals dating back to 2002 (in particular, see the October 2002 issue of the TDIL newsletter, available at ). The proposal, together with the supporting evidence, is rather large: over 500 pages. This makes it somewhat difficult to analyze the proposal but the following can be observed: - the proposal summary form indicates that 256 characters are proposed - the proposed set apparently duplicates entirely the existing Devanagari block - it adds coded characters for the consonants with an explicit virama, which can be represented today using a sequence of coded characters - there are numerous reserved code positions, with annotations of the form "Reserved for Kashmiri UU", suggesting that further research is needed - assuming an encoding model that extends the existing Devanagari block, rather than duplicates it, the number of proposed characters would probably be around 100. - while there is a commendable effort to document each proposed character, the form of the documentation and the resulting volume make it difficult to properly assess this proposal. In my opinion, the first steps ought to be: - to achieve consensus as to whether Vedic Sanskrit should be encoded entirely independently of Devanagari, or should instead be an extension of the Devanagari block - for the characters for which code positions are reserved in the current proposal, either research them and include them in the proposal, or ignore them entirely (with the understanding that they can be part of a separate proposal later) - find an organization for the supporting evidence that make it more manageable. I sincely hope that progress can be made on supporting the representation of Vedic texts in Unicode. ---