Name: rlT66838			Date: 08/15/97


1) Java uses Unicode for character encoding.
But, it is my impression that even in Java 1.1, 
Java relies on the underlying operating system 
to actually produce the glyphs and that 
different OS's support different subsets of 
Unicode.  This is contrary to the idea of 
write once -- run anywhere.

The solution is to license a Unicode font and
distribute it with the base Java distribution.
Hey!  Bitstream is giving away a Unicode font 
(Cyberbit).  Please license it from them and
make it a standard part of Java.  See
www.bitstream.com/cyberbit.htm

2) It is very difficult for users to access
Unicode characters.  At best, it can be done
with \uxxxx, but this is clumsy at best.  A
better solution is to support SGML public entity
names.  This is a superset of the familiar
HTML form: "&name;", for example "α" for
the Greek letter alpha.

Java needs to provide two methods: 

String entityToUnicode(String s) which converts
  any entity names in the string into Unicode
  characters, and

String unicodeToEntity(String s) which converts
  any non-ASCII character into its entity name.

These suggestions are relatively easy to 
implement and would really open up Unicode to
Java developers and users.

Thank you for considering my request.


======================================================================