My company uses an internally developed package to support internationalization/localization. However, it was developed some twenty years ago, and the libraries are restricted to one product line. I'm interested in where the state of the art stands. Is Unicode the base character set for all international efforts today? Do people still use gettext() and family? If using C++, should I be concentrating on the locale support in that language?
I've looked at the Wikipedia entry on Internationalization and Localization, which includes links to other sites and related topics. But there's a lot out there, and what I'd like is a source, especially a book, which serves as a good introduction to the topic on Linux/Unix in the current software scene. If the implementation uses C++, that's fine.
For example, years and years ago I read a book called Understanding Japanese Information Processing, by Ken Lunde, and much later, his revised book, CJKV Information Processing. Both were interesting, but of course focussed on Asian languages. Is there a book on the current art that is (spoken/written) language agnostic?