This can be confusing when you care enough about how the Han unified characters look but can't control the lang attribute (for example in Discord messages). Which is reasonable enough but can be surprising I think. lang='en' + system lang en -> 'Japanese' font
lang='ja' + system lang en -> 'Japanese' font lang='en' + system lang ja -> 'Japanese' font lang='en' + system lang en -> 'Chinese' font For example, on Windows 10+ with system font stack without system-ui:,
lang attribute and the system language are probably used to determine the fallback order. I'm pretty sure the font choosing algorithm takes into account the 'lang' attribute as well.
