Response to WG2 N1634, Incorporation of Bangla (Bengali)
Coded Character Set in ISO/IEC 10646

April 16, 1998

The Unicode Technical Committee generally concurs with the points raised

in WG2 Nxxx, contributed by Mr. Michael Everson in response to N1634, a

contribution from Dr Golam Mowlah, Director General of Bangladesh Standards

and Testing Institution (BSTI), dated 1997-06-29.

The repertoire of Bangla (Bengali) characters in the Unicode Standard

(which is consistent with ISO/IEC 10646) is based on the assumption that

rendering of text will be done by an algorithm, designed to provide the

correct conjunction formations from underlying component characters.

The key questions pertaining to the repertoire are:

* Is the character repertoire for Bengali script in both the Unicode

Standard and ISO/IEC 10646 sufficient for writing the Bangla (Bengali)


* Does every character of BDS 1520:1995 "Bangla Coded Character Set" have

an equivalent representation in the encoding of the Unicode Standard and

of ISO/IEC 10646, either as a single coded character or as a coded

character sequence?

Like ISO/IEC JTC1 SC2/WG2, the Unicode Technical Committee has a principle

of not changing character codes after assignment. Therefore, rearranging

the characters in the range U+0980 to U+09FF in order to match the

repertoire and order of BDS 1520:1995 is not possible.

The Unicode Consortium provides mapping tables for many national and vendor

standards, to define the relationship between the values in those standards

and Unicode values. These mapping tables are available on the Consortium's

ftp site, ftp://ftp.unicode.org. The Unicode Technical Committee would

welcome a mapping table for BDS 1520:1995 Bangla Coded Character Set, if

the Bangladesh Standards and Testing Institution were to prepare one.

If it can be shown, after careful analysis and taking the design principles

published in "The Unicode Standard, Version 2.0" into account, that one or

more characters in BDS 1520:1995 Bangla Coded Character Set cannot be

mapped, then the unmappable characters are candidates for addition to the

repertoire of the Unicode Standard and also to the repertoire of ISO/IEC


It should be noted that the Bengali script repertoire in the Unicode

Standard and ISO/IEC 10646 is intended to cover historic material and other

languages written in this script (e.g. Assamese). Therefore, characters

which are not used to write the Bangla (Bengali) language today may be