Here is an attempt to document what SEEMS to finally be working for me, to 
get unicode working with the JDK 1.1, and appletviewer.
Or if you are having problems with netscape's applet viewing on UNIX.

[This is for Solaris. Windows people, you may have to make your
  own adjustments. or see /java/unicode/mswin.html ]

In $JAVA_HOME/lib, edit the file "font.properties". It should already
exist.
[Or if you are using netscape to view an applet, 
   $MOZILLA_HOME/java/classes/font.properties]

Then add entries for serif, sansserif, monospaced, dialoginput,
but most importantly, dialog, in the following format:

   # "normal" full spec for font(s) of your choosing look something like
   #dialog.plain.0=-b&h-lucida *-r-normal-sans-*-%d-*iso8859-1
   #
   # Japanese font, presumably jis, here. This is a "lets grab something" game
   dialog.plain.3=*jisx0208.1983-0
   # And here's the "magic" that makes the above actually work
   fontcharset.dialog.plain.1=CharToByteX11JIS0208

 http://www.javasoft.com/products/jdk/1.1/docs/guide/intl/fontprop.html


The problem I hit here, is that there is a big long ugly section in
the docs, that throws sample CODE at you on how to write your own char
converter.  Not only that, but the major heading is 
 "Supporting User-Defined Characters"
Which is inappropriate when all I was trying to do was use the fonts that
come with my machine, not "define my own characters"

What it should probably say in big lettering, is 

 IF YOU USE A NON-UNICODE-encoded FONT (unicode fonts are very rare),
 then YOU WILL NEED TO SPECIFY "fontcharset" for that font, as in the
 example above.
 
 Sample values are:
    CharToByte8859-1
    CharToByteX11JIS0208

[and note that I don't see "CharToByteX11JIS0208" as an option, 
  documented at all]