Is encoding based on pronunciation a good idea?

No doubt all Indian languages are Phonetic! What is a Phonetic Language? It is faithful in making an EXACT representation of sound by the way it is SCRIPTED.
In other words, if you know how to "SOUND" a particular word, you know its "SPELLING" as well. It isn't true for English. For e.g. "DO" is spelled similar to "GO" but the pronunciation of both these words are totally different.

In almost all Indian language, it is almost predictable how a word is READ just by looking at how it is WRITTEN. In Computer terms, it is called as WYSIWYG = What You See Is What You Get

But is it 100% True for all Indian languages and for entire vocabulary?
Unfortunately NO!
I can site some examples but you can think of it also.

But the biggest mistake ever I would say for Unicode Encoding for Devanagari was basing it on the assumption that it is 100% True that led to non-standard workarounds in specific language issues.

all the half 'r' related descripancies are stemming from this particular issue.

No comments:

Post a Comment