YAPPARIGNAN -- Usage Hints

The input may be given in two forms. You can paste a Thamizh text from an existing file (TSCII fonts). If you use Keyman it may be possible to enter Thamizh characters directly. Or you may directly enter the text transliterated into roman alphabets. Care should be taken to distinguish nedils by using capitals or doubled small letters.

In the page "ceppiduvAy soltharum cIr" one may input a word, or a frag ment of a word, or manywordsruntogether, and ask YAPPARIGNAN to determine the type. (That is the input should be continuous without any space.)

The output will give details as nErniRainEr etc and the mnemonic vAypAdu (in this case kUviLangkAy). Also the fitness of that word as final cIr of a veNpA would also be given in the output. Of course mnemonic vAYpAdu will not be provided for cIrs of more than 4 syllables.

In the page for thaLai two independent words (rather cIrs) should be input in TSCII or romanized form. YAPPARIGANAN will try to describe the thaLai and also classify it. Unclassified thaLai combinations (for example 'viLam mun niRai' is a persona non grata in all types of pAs) will cause YAPPARIGNAN to get into humorous mood resulting in some facetious messages. YAPPARIGNAn is aware of terminal ugaram doing a vanishing act in the presence of a vowel at the start of next word. It behaves intelligently, and gives correct thaLai, taking into account this phenomenon.

For classifying mOnai one needs to type an adi consisting of 4 cIrs or 3 cIrs. An adi of a viruththappA will be safely ignored after the 4th cIr. YAPPARIGNAN looks at the matches/similarities in the first letter and classifies. Whereas YAPPARIGNAN is smart with thaLai, its weakness is exposed when determining mOnai. Certain matches will be missed out if the text followed 'puNarcci rules'. For example:

"anbu maRanum udaththAyin ilvAzkkai" In this the 'm' at the end of first cIr joined with following cIr's starting letter thereby obscuring the mOnai.

Or it can wrongly see a mOnai where there is none. "kurangkuthAn avanmukaththiR koppum" Here the 'ku' at the end of the second word got tied up with the beginning of the last cIr tricking YAPPARIGNAN into believing it is a mOnai.

Another problem is aLapadai. Ideally it is better to input it with "udambadumey" to avoid too many vowels crowding at one place. A 'v' or a 'y' would do, though it is not very pleasing. To be more precise the problem is the occurrence of uyir ezhuthtu in the midst of a cIr. The well known thrukkuRal, "iniya uLavAka ..." needs to be input with care. The two words "kani" and "iruppa" which ocur together as one cIr, should be entered as "kaniyiruppa". Without that 'y' YAPPARINJAN will think there is a nedil in that word.

For analysing a full pA, at present the user has to say whether the checking is for a veNpA or for Asiriyam or Viruththam. System does not automatically determine that. Also for processing viruththam the user has to specify how many cIrs per adi. The advantage of this is, we are free to start a new line after 3 cIrs without worrying if YAPPARIGNAN would consider that as an independent adi. At present there is no mOnai checking for viruththam. And "Centhodai" merely refers to lack of mOnai, and YAPPARINJAN does not try to understand any meaning or irony in the pA! And the system cannot ignore consonants while parsing ("mey nIkki alakiduthal" is not done)

AikAram and aukAram are treated uniformly as kuRil.

Books used for grammar definitions.
1. Grammar textbook for class 9 and 10 by Tamilnadu textbook Society.
2. "Kavinjaraaka" -- A.ki. Paranthaamanar, First Edition 1963., Paari Nilayam, Chennai