Werner Lemberg wrote:

> I think he has rather meant 16bit character support...

In fact, it means 32bit support. Currently there are 16 additional planes
(in ISO 10646 speak) assigned for the allocation of more characters.
Unicode 2.0 handles this problem by the surrogate mechanism (which is
aequivalent to UTF-16 of ISO 10646), meaning that a pair of two 16 bit
characters represents one 32 bit character out of the 16 additional planes.

Alternatively, it could mean UTF-8 support (32bit fixed width characters
are transformed into variable length sequences of 1--6 octets). This can
be handled with TeX classic on the input level. Of course it can't be
handled with TeX classic on te internal or output level, since larger fonts
are desparately needed.

--J"org Knappen