Doesn't quite work IMO. What if, rather than getting either full points or nothing, you were scored based on how close your guess was (in terms of language families) and how much text you needed to look at?
Thanks for your comment! I have been troubled by the first opinion, as it is difficult to score "linguistic distance" quantitatively (like GeoGuessr) even in an academic setting. The second opinion is a good idea, and I will add it to my TODO list!
You can basically guess the language based on the script in which it's displayed. It would be better if the language choices were displayed as a translation to your native language. So instead of 日本語 it should just say "Japanese"
I think that's the idea. Not much different than differentiating e.g. German and Italian as they use mostly the same alphabet.
For what it's worth, I can identify Farsi from Arabic as I can read Arabic and Farsi has some additional letters. I suppose that German and Italian would be a similar situation for most HN readers.
Presumably if you read a third language written in the Arabic script (e.g. Urdu or Pashto) then you could probably tell them apart without necessarily being able to read either.
After all, you can probably tell German and Polish apart without being able to read or speak either one.
> You guessed English (en) but the right answer was Simple English (simple).