Around 8 o'clock on Sep 30, Tomohiro KUBOTA wrote:
> In future, every softwares must be multilingual. To achieve > this, future API must prohibit apps-developers to write non- > multilingual softwares. In other words, future API must not > provide a way to write non-multilingual softwares. While I agree with the goal of multi-lingual software, the reality is that developing true multi-lingual software is significantly harder than focusing on a single language. I'd like to develop a single interface that can support both multi-lingual and uni-lingual applications. This is the primary reason I've resisted implementing mb/wc interfaces for Xft. By having application developers confront Unicode at the outset of development, there's a much greater chance that they'll choose to store data inside their application in Unicode form. Suddenly, the conversion of their application from English-only to multi-lingual becomes one of adjusting the input/output routines and not one of completely restructuring the internal data structures. If we're really lucky, they'll choose UTF-8 so that when the OS finally supports UTF-8 locales, the conversion routines become no-ops. Of course, developers can continue to ignore the problem and convert a fixed character set to Unicode for Xft, but at least they know they're taking a shortcut. There's no way to force developers to do the right thing, we can only try to persuede them gently. A confrontational approach with mb/wc interfaces would just keep them using core X fonts and their native encodings. [EMAIL PROTECTED] XFree86 Core Team SuSE, Inc. _______________________________________________ I18n mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/i18n
