Linux is not a static system: it is a system that is changing and maturing rapidly. The state of Unicode and internationalization in Linux today reflects this condition of rapid growth in both positive and negative ways.
On the positive side:
① The Open Source development process creates an open dialog between developers and users which results in rapid progress. We saw an example of this with the Unicode editor Yudit which facilitated the efforts of the Bengali Linux localization project.
② A solid Unicode-based infrastructure and powerful applications are available to both developers and end users at minimal cost or nearly free. A professional font design program for Windows or Mac costs about U.S. $ 349.00, or nearly 1/5 the annual salary of a person in Bangladesh. In contrast, George Williams' FontForge program is free and yet has a sufficiently powerful feature set for professional work.
③ Essentially all configuration files, locale files, message translation files, keyboard maps, and input method maps are in plain ASCII text, plain UTF-8 text, or XML formats. This makes Linux systems easy to customize using a simple text editor like vi or Yudit.
On the negative side:
① There are too many complex text layout engines. It would be better if efforts were unified in the design of a single, toolkit-independent library following the examples provided by FontConfig and FreeType. Will the “FreeType Layout” library (not released yet) emerge as a future contender for the CTL title?
② Software installation on Linux is still too difficult for the average user. In real life, the RPM software package system used by Redhat and SuSE doesn't work as well as some competing systems (Debian's apt-get, for example). Compiling from source works best, but some “user” distributions don't have developer tools installed by default.
③ Although Redhat and SuSE now use UTF-8 locales by default, many of the other Linux distributions still default to legacy encodings. This is an unecessary obstacle.