On Friday 20 September 2002 11:27, you wrote: I want to be sure that any text files I'm working with will be understood by programs such as xsltproc, and Java apps, these apps correctly understand them as UTF-8. I believe there may be a way to declare a list of priorities. After I get some sleep, I'll look into it. Well we can't steel their font images, but we can draw pictures of them: http://www.unicode.org/charts/ Hard copy. It's Latin Extended B 01EA, and 01EB. You can find it on the above linked site. There are fonts for for most characters I want to use. I have input methods for many of them, but these are not necessarily pleasant. I've found looking at the following files to be a bit instructive: http://66.92.149.152/open-source/org/xf86/en_US.UTF-8/ http://66.92.149.152/open-source/org/xf86/en_US.UTF-8/Compose http://66.92.149.152/open-source/org/xf86/en_US.UTF-8/XI18N_OBJS http://66.92.149.152/open-source/org/xf86/en_US.UTF-8/XLC_LOCALE They are off my SuSE box. I'm pretty sure they are the prodocut of XFree86. Mozilla does a pretty good job with these. XEmacs handles many of the fonts, but it requiered a bit of persuasion. -- STH