Salaam Mohammed, > Please understand that this has to be easy, I > can't imagine a typical typist > having to type 'logical' codes as if he is a > programmer!
Typing logical codes is already something that is imposed by the Unicode standard. A very good example is the ZERO WIDTH NON-JOINER (U+200C) used in Farsi for numerous words such as "mi-zanad". The Farsi typist has to enter this logical zero-width non-joiner code in between "mi" and "zanad" otherwise the word doesn't render correctly. Fortunately this logical code has been conveniently integrated into the new standard Farsi keyboard as "shift+space". You can read more here: http://students.washington.edu/irina/persianword/zwnj.htm So basically there are cases where we need to use logical codes because the other alternative (one unique code for every combination) can become redundant. The input of these logical codes can be simplified for the user at the keyboard layer. For example, for the sequential fathatan the user can enter "shift+fathatan". Similar for other variant dammatan's, and kasratan's. > But dammatan looks very different from > damma+damma. As far as I know, damma+damma is never used in Arabic in a sequence to represent two dammas next to each other coming after a base letter. Damma, fatha, and kasra by definition associate with only "one base letter". So having two seperate dammas for one letter is not used in Arabic. That's why Tom is suggesting that this un-used code sequence can be declared canonically equivalent to a dammatan. Likewise kasra+kasra and fatha+fatha can be declared canonically equivalent to kasratan and fathatan respectively. Of course these canonical equivalence definitions, once accepted, would be part of the Unicode standard, so it wouldn't be just our way of doing it but the standard way of doing it. Regards, Mete _______________________________________________ General mailing list [EMAIL PROTECTED] http://lists.arabeyes.org/mailman/listinfo/general
