While nowadays the main languages (Mandarin, Japanese, Korean) in East Asia are quite well supported on computers, some peoples native languages were left behind. This is true for Taiwanese Minnan and Hakka, two languages spoken by 70 and 30 percent of the population in Taiwan respectively, as well as for many other minority languages throughout East Asia.

Until now, people who want to use Taiwanese Minnan or Hakka on their computers, have to buy commercial software solutions to solve this problem. But these commercial solutions introduce another problem: incompatibility. Each software vendor promotes his own solution, but documents typed with one program cannot be displayed properly with other applications, let alone by users who don't have one of these special programs installed.

This problem has a reason: up to now there exists no standard in Taiwan or elsewhere, which covers this issue and provides input tables or character sets, which contain all necessary characters and their input sequences.

While most Hanzi used in Taiwanese Minnan or Hakka are already in the Big5 standard, many characters are still missing. Even the CNS11643 standard does not contain all characters necessary to type or display Taiwanese Minnan or Hakka properly.

In the past years another international standard has emerged: Unicode. Unicode has been designed to support all languages and their characters in the world. But even there many characters necessary for Taiwanese Minnan and Hakka are still missing.

Consequently, there does not exist any "free" font, which covers all those characters, nor does there exist any input method to type those characters.

The commercial applications I mentioned, solve this problem by inventing their own standards. This however leads to the problem, that users who don't use one of these programs, can neither display nor type Taiwanese Minnan or Hakka properly and document exchange is highly complicated if not impossible.

This is one of the reasons, why I started the CJK-Unifonts project.

This project has the following goals:

  • create a "free" set of Truetype fonts, which contain all characters necessary to display Taiwanese and Hakka.
  • for those characters, which are not yet included in the Unicode standard:
    • identify the missing characters and
    • submit them to Unicode for inclusion into the standard.
    • until those characters are included in the Unicode standard, use Ideographic Description Sequences (IDS) to describe the glyphs.
  • create an input method for typing Taiwanese Minnan and Hakka. This input method:
    • can be used on Windows, Linux, Mac OS and many other systems
    • can take input in Zhuyin, Tongyong Pinyin, POJ and maybe other methods
    • can output Hanzi, Zhuyin, Tongyong Pinyin, POJ and maybe more forms (e.g. Hanzi with Zhuyin attached)
    • can lookup a dictionary to make autocompletion or auto selection of words possible.
  • both, fonts and input method can be used on any Unicode supporting system (i.e. Windows 2000, XP, Vista , Linux, Mac OS and others) Especially for Taiwanese Minnan and Hakka Hanzi, there is a lot of confusion among students and users, because publishers often use wrong characters as they currently cannot type or display the correct ones.

In this project we work closely together with respected dictionary authors and researchers to assure that the dictionary and input table entries contain correct Hanzi.

This is especially important for teachers and students, because if every school uses different teaching material, different romanization systems and different Hanzi to teach Taiwanese or Hakka, written communication among students, as well as future publications will suffer from the same problem we have currently: that not everyone can read and understand them, but only those people who have learned the same characters or romanization systems.

This project aims to create a foundation for better written communication and teaching material, through standardized characters, vocabulary and input methods.