unicode mystery

[Alex Rice]

Ignacio Vazquez-Abrams wrote:

> Something that can be done is that a new Unicode codec can be written for an
> encoding like 'MSUnicode' that takes Microsoft's non-standard Unicode encoding
> and transcodes to or from standard UCS-2. Take a look at:
> 

That sounds like an ideal solution. However in general, characters some
through from Word just fine as Unicode. In fact, Word's "Insert Symbol"
font browser seems to have a unicode-ish breakdown of the character
subsets. I'm guessing Word uses unicode internally. 

However, for the Symbol font, something very very screwey is going on- I
can't even reproduct the behavior I first posted about. In fact the
behavior is different, but still inconsistent in other documents I have
created with Symbol font. It's very strange.

Thanks,

Alex