At 12:11 01/11/07 -0500, Jungshik Shin wrote: > > I immediately think > > of text with more than one language being displayed. > > In this case Han disambiguation is a problem and some > > mechanism is needed Or eLsE ThE ApPlIcAtIoN WiLl gEt > > tHe FoNtS WrOnG WhIcH WhIlE It iS ReAdAbLe iS CoNsIdErEd > > uNpLeAsAnT. Just like ascii is readable (and spell > > checks) even if the case is wrong. > > Well, it's not that serious *unless* a program *mixes* >*multiple* fonts to meet demands for glyphs(a single (CJK) font can't >provide all the glyphs necessary and a program has to look for glyphs >in other (CJK) fonts with very different look and feel). That is, if a >program just uses a single font or a set of *consistent* fonts (that get >along well with each other) to represent *all* Han characters, it's not a >problem. Using Taiwanese glyphs to represent mixed trad. Chinese/Japanese >text(say one paragraph is in trad. Chinese and the other is in Japanese) >for Taiwanese users are perfectly all right in *editors* (although maybe >not in a word processors). > > It gets very ugly if multiple fonts from different sources with >very different design principles and goals are mixed together to represent >Han characters in a given run (e.g. paragraph) of Unicode text.
This is an excellent summary of the (non)problems with cjk unification. Thanks! Martin. _______________________________________________ I18n mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/i18n
