[UTF-8] Aspell and UTF-8/Unicode

[Danilo Segan]

Hi Kevin,

Kevin Atkinson <kevin@atkinson.dhs.org> writes:
>
> Unless you know how Aspell works internally, don't tell me how to write a 
> spell checker.
>

Sorry you got my message this way -- it was never intended to be
insultive (which is how it appears you got it).

I also never touched on spell checker specifics, but only concerned
myself with 8-bit character sets vs. UTF-8.

> If you care to educate me on something you wish to spell checker that 
> Aspell can't handle I am all ears.  It may be that Aspell currently can 
> not handle the task even if it used Unicode internally.

Accented Cyrillic.

Serbian Cyrillic and Latin in one dictionary.

I'll be happy to provide concrete examples as well.

Sure, it is already possible to have a separate Cyrillic and Latin
dictionary, and use that.  But it's about (in)convenience -- it is
already possible to hire a proofreader, so why bother with aspell?

> Are you saying that as long as Aspell uses 8-bit internally it will be on 
> the "Bad Software" list, even if this is transparent to the end user?

Nope, of course not.  But it is not transparent to the "end user",
because end user cannot do expected stuff with it (depending on what
end user might want to do, but I apparently cannot get the above
done).  I don't enjoy having as much software on Bad Software list as
possible -- I'd much rather enjoy that list to be as short as
possible.  The idea here was that having such a list might help
shorten it.

Cheers,
Danilo