This sounds plausible. Unicode as used and implemented on modern OS has a nasty quirk that texts become patchy mix of the language in use + random Simplified Chinese equivalents(apparently the opposite still randOmly haPPEn in Simplified Chinese systems too - showing Japanese pieces out of nowhere). The official Unicode Consortium sanctioned solution to this problem is to specify and switch fonts wherever and however appropriate, even mid-sentences, which isn't a great solution, if not unreasonable for a lot of developers.

Creating language-specific fonts that can be just forced everywhere to eliminate random pieces from other languages solves this problem. At least everything will be consistent.