Re: Tamil Collation vs Transliteration/Transcription Enc

From: Richard Wordingham (
Date: Fri Jun 24 2005 - 20:25:13 CDT

  • Next message: Asmus Freytag: "Re: terminology: plaintext (was Re: unicode Digest V5 #149)"

    Sinnathurai Srivas wrote:

    > However it is important that we do not add any Grantham letters to Tamil.
    > For example we are considering deprecating the new 0bb6 sh addition.

    Who's 'we'?

    > Tamil is an ancient language with sophisticated Grammar (probably the
    > oldest Grammar in the world) to back it up with. grantham does not base
    > the writing system to this Grammar. It would be un acceptable to add any
    > more Grantha to Tamil.

    It makes sense that the *extra* Grantha letters are not needed for Classical
    Tamil. Modern Tamil has sounds that Classical Tamil did not.

    > Unicode encoded 0bb6 without proper consultation, again offending the
    > language in a serious way.

    Tough. It was still being used.

    > As for aspirated, i mean dh = d+aytham, kh=k+aytham
    > For example unused code point 0b96=0b95+0b83. கஃ (KH)
    > Tamil also defines 0b83+0b95 ஃக (HK)which is not find in Grantham.

    However praiseworthy, digraphs like these do not belong in a truly phonemic

    > In any case it has been a long and hard struggles to keep Grantham out.
    > Ofcourse Grantham has won over mostly all of the Indic languages. Tamils
    > wish to keep it the way it is defined in the ancient and sopisticated
    > Grammar. I do not think Uniocde with its power would once again start to
    > distroy this.
    > I was talking to some one about adding the special symbol "th" to English
    > encoding and he said Unicode does not have the power to arbitarily change
    > English, even if we provide amble evidence of usage in English.

    Even before Unicode, ISO-8859-1 enabled the restoration of the two Old
    English letters (U+00F0 and U+00FE, plus their upper case forms) for the
    'th' sounds, which had never been lost from Icelandic (which basically got
    them from Old English). There's already a special 'th' symbol compounded of
    't' and 'h' encoded - U+1D7A LATIN SMALL TH WITH STRIKETHROUGH, and there's
    only the UTC to stop the 'th' symbols from Pitman's Initial Teaching
    Alphabet (i.t.a.) being encoded.
    However, English would only be changed if influential enough people chose to
    use them.

    > But he also said Unicode has the power to change Tamil and other languages
    > as there is no significant power that can stop Uncode, if Unicode decides
    > to do it.

    > I hope things do not go that far and hope Unicode will help to deprecate
    > 0bb6, as this is an addition not necessary for Tamil. It is not justified
    > to attack power less people with great and classical traditions, because
    > one has power.

    Similarly for Tamil - adding (or restoring) new letters only affects the
    language if Tamils then choose to use them. What Unicode might do is to
    stop letters being borrowed between Indic scripts. (Uniscribe would be the
    point where problems occur - rather as the version on my machine insists
    that aytham is not a base letter - or is this particular behaviour governed
    by the font?) Unicode, especially in combination with Uniscribe, may also
    inadvertently help 'oppress' groups by denying them the letters they need
    for their traditions, e.g. extra Grantha letters for Sanskritising Tamils or
    the 'redundant' letters of Lao for those who want to write Pali in their
    first script. (The latter group may not exist - I gather that Pali is
    normally written in the Tham script in Laos.)

    Incidentally, is there any way of co-ordinating the squatting of deleted Lao
    consonants in their rightful places? I thought the appropriate illegal
    encoding was fairly obvious (Lao = Thai + 0x80), but I was confused by a
    font that thought that one squatting place was as good as another.


    This archive was generated by hypermail 2.1.5 : Fri Jun 24 2005 - 20:37:50 CDT