P11: Assamese letters sort order

Last updated: August 31, 2004

In Assamese, the letter U+09F0 BENGALI LETTER RA WITH MIDDLE DIAGONAL sorts between U+09AF BENGALI LETTER YA and U+09B2 BENGALI LETTER LA (U+09B0 BENGALI LETTER RA is not used). And the letter U+09F1 BENGALI LETTER RA WITH LOWER DIAGONAL sorts between U+09B2 BENGALI LETTER LA and U+09B6 BENGALI LETTER SHA. Thus, sorting on code points is not correct.

: TDIL proposes to deprecate the existing characters, and to re-encode them so that code point sorting works:

It is a long standing position of the Unicode standard that sorting by code point order is not a viable goal.

Furthermore, it is far more damaging to the standard to move (or deprecate/reencode) characters than to have to use a more or less sophisticated sorting strategy, such as the UCA.

The default collation table for the UCA already sorts Assamese correctly. From http://www.unicode.org/Public/UCA/4.0.0/allkeys-4.0.0.txt:

Document History