Hello Naena Guru, am 2012-07-15 20:36, schrieb Naena Guru:
my challenge stands [...] to show how romanized Singhala violates any standard in what specific way.
Your “Romanized Singhala” neither complys with, nor hurts any standard. As it is your own invention, it is simply out of scope of all existing standards. However, tagging your “Romanized Singhala” as ISO 8859-1 encoded text, certainly violates several standards! E. g., ISO 8859-1 defines code 41 (hex) to be a Latin character (viz. “A”); so this code can never represent a Singhala letter, when it is part of an an ISO 8859-1 encoded text. Likewise, your scheme violates all pertinent standards of text tagging, such as RFC 2046: All of those describe how the encoding of a particular data stream is specified, so the receiving side will interprete it correctly. Clearly, if you send a Latin character (and tag it as such), you cannot expect the receiving side to interprete it as a Singhalese character. So, the only way for your encoding scheme to be exploited without violating the pertinent standards would be to register it under a new name, and then tag your data accordingly. However, as there are apparently no technical problems with the existing solution (i. e. properly tagged Unicode data), your new encoding scheme will properly not gain wide ac- ceptance. You can serve your language much better when you try to improve current solutions within the realm of existing standards. E. g., you could point out errors (or shortcomings) in existing fonts, editors, keyboard drivers, and other software and suggest (or even provide) better solutions. Or you could publish tutorials or examples of good practice. In any case, it would be wise to know the existing standards and comply with them. Best wishes, Otto Stolz