In order to allow reliable and consistent support for non-English languages, a consistent encoding needs to be used.
- UTF-8 encoding should be mandated for Java source and textual story files.
- ISO-8859-1 with Unicode escaped characters for locale resource bundles.
The requirements need to be clearly documented, with links to tools that allow conversion of native to ascii based unicode, tools that can allow users to easily and efficiently specify keywords in their locales.
A standard set of locales can then be supported and bundled as core resources, so that users need only specify new LocalizedKeywords(<my Locale>).
We may at a later state a further layer of conversion using some of these tools.