Re: Unicode implementation concept

Mark Evans wrote:
> ...  The low-frequency
> argument is one I've already made myself.  It tends
> to mitigate the serialization problem, the only
> real drawback.

Mark and I showed a list of further drawbacks.

> For UTF-8 in various Asian and Indian tongues
> the frequency of wide characters is heavier.  In

No - "wide" characters in this sense are used in UTF-8 for _everything_ outside of English. Multiple 
UTF-8 bytes are required for Greek, Cyrillic, Arabic, Hebrew as well, and also for "real" 
punctuation and accented Latin letters.

> these languages the performance gain from fixed
> code point indexing is obvious, for slicing and
> dicing purposes.

Again (and again and...), only if code point indexing is used. This may be the case for your "new 
language", but it is not in normal software development. There is normally no reason to index code 
points instead of code units. If you choose to do everything differently from everyone else, then 
that is your choice, but not something inherent to string handling in general.

You will probably find a more productive discussion forum on a mailing list and newsgroup that is 
dedicated to your "new language" and where the participants are familiar with the choices made there.

Best regards,
markus

-- 
Opinions expressed here may not reflect my company's positions unless otherwise noted.