Hacker News Reader

Same codepoints. Software can only choose the right font if it knows it's in the Japanese language. I'm sure if you set your browser's language to Japanese it would select the right font, but since it's probably set to English instead, it's selecting a "default" assumption for the language of those glyphs.

App localization (where you specifically select "Japanese" as the language you want an app to display in) is the main target of this article, I think.

View on HN

Linguistic identification can in principle be a bit smarter than that. Aside from the obvious "if it has kana it is Japanese" metric, there are also differences in code point choice which differentiate Simplified, Traditional, and Japanese. Fully accurate identification is certainly expensive, but quick early-out heuristics could be developed.

This reminds me of guessing codepoint likelihood based on character distribution.

indeed it could be better, though do you know of any apps that actually try to be smarter?

In principle this is a job for Unicode font rendering libraries and/or standard OS text widgets, not app developers.

Reply on news.ycombinator.com