Paul's recent mail and Matz's response explain much of
the proposed new unicode behavior -- for me at least.
Thank you both.

I've got a couple questions to add to Paul's list:

When a string's encoding is UTF-8, will it be represented
internally as an array of bytes (as v1.6/v1.8 Strings are)?

Will strings having different encodings be instances of
the same string class?

  utf_8_StringInstance.type == utf_16_StringInstance.type

Or, will there be a heirachy of string classes for
specific encodings all of which inheirit from class String?

-brent