Re: Offtopic : Unicode and Bengali

From: Martin Kochanski (unicode@cardbox.net)
Date: Tue Mar 05 2002 - 06:09:45 EST


Now it may be pedantic of me, but I can just imagine a defence for the approach that these guys are taking.

The correct way of implementing Bengali is (as I think you are saying) as follows:

1. Use the defined Unicode code points.
2. Use (or design) a font that is intelligent enough to combine these code points into conjunct glyphs.
3. Do all this on a computer whose rendering engine is intelligent enough to understand and handle the font.

Once 2 and 3 are so universal that you don't need to think about it, akshor.com's approach will be obsolete.

But (2) the design of an intelligent font is quite a bit more difficult than designing an unintelligent font that contains a lot of private use characters. Perhaps there are lots of good Indic fonts around already: I don't know.

And (3) it is not yet realistic to expect that all software will be able to handle the non-1-to-1 relationship between characters and glyphs, and it won't be for some time. You need an application that understands the distinction; an operating system that supports it properly; and display and printer drivers that don't corrupt the result.

So as an *interim* solution, I think that this system has merit. I agree with you that it's not clear that they think of it as an interim solution, so they do need to be educated... and they also need to provide a transliterator that will take you back from their private-use character set to real Unicode, so that when their system does become obsolete, people don't get stuck with meaningless data.

At 09:09 05/03/02, Dhrubajyoti Banerjee wrote:
>Hi,
> I was taking a bit of time off by searching what is available for Unicode
>on Indic Scripts. I came across this site.
> http://www.akshor.com/
>
>Although I was quite amused at first, with excerpts like, "...To minimize
>the codepage issue a brand new technology has lunched just a few years ago.
>", I understood that people still do not (or do not even try to) understand
>the basic philosophy behind coding standards.
>
>For example:
>"..In ISCII and Unicode, most of our conjuct characters are missing and
>another complexity in the input method; to form a 'reph' above the letter
>'ba', for example, it is necessary to type: 'ra hosonto ba', the software
>will in turn then convert this into: 'ba with reph above'.
>I thing we need not be restrained by these so-called 'standards'. Because,
>they can't and will not serve our need (Bengali) in my humble view. Thats
>why we toke this project at our hand and working to impliment a universal
>input method.."
>(http://www.akshor.com/project1.html)
>
>And
>
>"..Unicode Consortium assigned 0980-09FF for Bengali (see previous page).
>But this is not enough to saperate all our characters. Thats why we used the
>Private Used Area (E100-E4FF) of Unicode for our project to assign &
>saperate all our characters (excluded in UCS) including extended
>characters/glyps and symbols. "
>(http://www.akshor.com/project2.html)
>
>In fact they have put a huge number of "conjunct characters" in the Private
>Use Area. Its a pity because it means that so many people still do not even
>understand the difference between characters and glyphs.
>
>I hope Unicode proliferates fast in these areas so people can understand it
>and use it without wasting time in such activities as reinventing the wheel.
>
>
>regards,
>Dhruba
>
>_________________________________________________________________
>Chat with friends online, try MSN Messenger: http://messenger.msn.com
>
>
>
>



This archive was generated by hypermail 2.1.2 : Tue Mar 05 2002 - 06:27:34 EST