Roozbeh,

 > On Tue, 13 Feb 2001, Marcel Oliver wrote:
 >
 > [Regarding UTF8]
 > > - Diagnostic messages could (although not with current TeX engine) be
 > >   output in the correct script.
 >
 > Really? Not with current TeX engine? Why?

it is not possible for log entries regarding the typesetting progress, eg
overfull box indications will be presented in the encoding of the font (which
in itself is not wrong but doesn't necessarily help if yo try to find that
text in your source:-)

and it is also not possible really in diagnostic messages produced from the
format or packages since the script that is used for them is hardwired into
TeX and at most changable at invocation, ie it is some 8bit thingie and might
in fact represent the upper part of 8bit as ^^ notation (depending on the
implementation)

as I understand the situation, Omega has the same defects though they appear
slightly different due to a different internal model (see the longer mail
about LICR and OICR)

 > > - The "combining characters" of Unicode are difficult to handle with a
 > >   TeX based parser.  (Does "difficult" mean "impossible to get
 > >   right"???  What are the issues???)
 >
 > Every letter should be made active to look forward to find the combining
 > character sequence after it, and then puts that over its own head! I don't
 > think this is impossible, you need to loop until a non-combining char is
 > found.

David explained what that would do to tokenisation of \begin etc (six tokens
instead of one), but yes you can provide an surface interface that would work
in this way. Only it would make LaTeX a lot lot slower without any benefit for
the majority of users (which goes back to my point of it being impossible to
make such a change if there aren't any real cute features those people wish to
have, to make the overlook other changes)

 > > - The output encoding is limited to 8 bit fonts, which may not be
 > >   enough to get correct kerning for some languages. (Can someone
 > >   confirm or correct this???)
 >
 > We need some examples. I can't find any.

Greek might be one if you require (as LaTeX currently does) that visible ascii
is part of the font encoding.

frank