Friends,

I have uploaded a proposal for the primary character names in the
files section under "tscii char name proposal-rev5.txt".

OLD ASSUMPTIONS:

1. Char names are transliterations
2. IETF requires the char names to be identical to Unicode char names
3. Only one char name should be given for each character/glyph

RESULTS OF ANALYSIS AND RESEARCH

1. Char names are not transliterations.

Unicode char names for Tamil are not meant to be used as
transliterations. They don't follow any recognized transliteration
standard. They are all in a single case, do not use numbers or other
signs and they cannot disambiguate between mey letters (consonants)
and uyirmeys (consonant+vowels) if used as transliterations.

2. Char names are names of characters that don't follow any
transliteration

This is obvious by looking at the unicode char names Aytham,
Visarga, Virama, as well as the names for single and double
quotation marks in the TSCII specifications. The character names
also don't follow any transliteration standard. For instance, the
name for the Aytham letter is spelled as "AYTHAM" and not "AAYTAM"
as would be expected from adherence to Madras Tamil Lexicon or other
transliteration standard. Instead it uses the common names for the
letter.

3. IETF does not require the char names to be identical to Unicode
char names

Though this is true by examining the registered char set names, it
makes it easier for programmers familiar with the Unicode chart to
see identical names in the TSCII.

4. Two sets of char names have been given for each character/glyph

This is obvious from the Sinhala Unicode standard

5. This proposal does not need to challenge, amend, or reform any of
the transliteration standards, conventions or proposals.


RATIONALE FOR THE PROPOSAL:

In several non-Latin character sets such as Greek, Hebrew, Arabic,
Sinhala and Tamil, characters have names. No Greek language
encoding standard would even consider using A, B, G, D, instead of
Alpha, Beta, Gamma, Delta, for the character names. Similarly, a
Hebrew standard would be expected to use names like Alef, Bet,
Gimel, Dalet and an Arab standard would use Alef, Beh, etc. The
Sinhalese standards were careful to include the character names
Ayanna, Aayanna, Aeyanna, etc. It is an oversight of the ISCII
based standards to have missed the character names. Thus, the ISCII
Tamil standard and the Unicode Tamil standard that derives from
ISCII have neglected to include the Tamil character names.

It is this oversight, this proposal wishes to correct.

Several of these character names are used as is in the Tamil grammar
books and other reference works dating from Before Common Era and up
to 2003. Tolkappiyam (200 BCE - 200 AD) the ancient classic,
Nannuul (13th Century CE) and until the 19th century the ultimate
standard Tamil grammar, grammar books by Suddhanandha Bharathiyaar
(1950), Dr. Paramasivam (1995), Singapore Siddharthan (2003), and
general books on Tamil language and linguistics by Dr. Mu.
Varatharasanaar, T. P. Meenatchisundaranaar, are among the several
books that support the notation in this proposal.

BRIEF ANNOTATIONS ON THE PROPOSAL:

I have proposed the character names in consultation with noted Tamil
scholars, Prof. George Hart, holder of Tamil Chair at University of
California at Berkeley and Prof. Ilakkuvanar Maraimalai, Depart of
Tamil, University of Madras, Tamil Nadu.

Essentially, it follows the standard described by TolkAppiyam (200
BCE - 200 CE) in adding -karam to the short vowels (kuRils) and -
kaaram to the long vowels (nedils). The pure consonants are
referred to us -kara mey as in "kakara mey" etc.

I left the Grantha characters without any change except to note that
they are uyirmeys or pure consonants. I couldn't find any reference
where they are also modified with the -karam/-kaaram suffix.

I have used common names for the characters such as Paththu rather
than pattu that would be compatible with the MTL transliteration
notation.

I have left the retroflex N as NN without any modifier. However, to
distinguish between the dental N (thannakaram) and the alveolar N
(Rannakaram), I have used the linguistic terms dental and alveolar.
I have tried not to use any transliteration notation to distinguish
between these two Ns.

I have included the Unicode Tamil character names as secondary names
without any change except to use the latest notations such as Aytham.

Please go over the proposal and send your feedback including typos
etc. I have already discovered that I have spelled PATHTHU AS
PATTHU. I am sure other notations need to be carefully proofed.

Thank you for your consideration.

anbudan,

Mani M. Manivannan