The missing info: Has Duolingo user growth increased roughly uniformly across languages, or did certain languages drive the growth? I could easily tell an alter

The missing info: Has Duolingo user growth increased roughly uniformly across languages, or did certain languages drive the growth? I could easily tell an alternative story about language learning becoming popular recently. As a recently "acquired" paid user, I have little good to say about the kind of gamification that DL is using, and the only reason for my use of the app is actually trying to learn the basics of a language. As a game, it is unconvincing. The exercise selection mechanism looks reasonable but has some easily fixable lapses (ambiguous English words like "you", "moose" or "deer" when singular and plural both make sense). The gamification is cringe: mechanics mindlessly copied from successful games without understanding what makes them work and why. Character progression is mostly linear. The multiplayer features are... mainly comparing your XP with friends. Useless interrupts and load screens get annoying after the first 10 minutes. I would never play DL just for the sake of play. The reminder emails? I don't need them, but I can imagine them being useful. It seems that educational games are one of those fields of gold that are still around in computing, and there are still low-hanging fruit left to be picked, particularly as more and more grading can be done by AI. So I am not surprised that products with obvious flaws and general lack of game design sophistication can still be highly successful and useful. But one day, someone will do things right and DL will be forgotten. Imagine a browser plugin that helps you read a text in a language you're trying to learn. You get points for correctly translating a sentence, proportional to the complexity of said sentence. You lose points for mistakes, and you can spend points for having the computer translate a word for you. Most of this should be pretty easy to implement, except for the part that would check a translation for correctness, which might need some AI that isn't yet in consumer land.