I think that there are a lot of issues with the way Duolingo motives people through unhealthy incentives, but research does show that it works. Though it will need to be used in conjunction with other methods of learning to be truly helpful. But then, why would you learn a language if you’re not going to actively use it?
That research you link has been conducted by Xiangying Jiang and Bozena Pajak (first and last authors). If you want to contact them about how totally objective and unbiased they are, here are their emails:
That's too dismissive in my view. Duolingo isn't as good as in-person lessons, but it's still way better than nothing. I'd also argue that grammar exercises are cripplingly boring for most people, so if the choice is not learning or grammar exercises they'd choose not learning, which is worse. I'd also argue that you're only talking strictly about learning language in the most efficient way, but missing out on the fact that if you can make it fun, the enjoyment in itself is a valid reason for doing something.
Edit: yes, I am dismissive because I've tried it and seen that its obviously a scam, in the sense that it's sold as a fun way of learning but it's actually just a fun way of not learning.
Duolingo is extremely passive. you're not learning a language, you're learning how to answer correctly on an app.
It's a lot more time efficient to do exercises from a grammar. But... It doesn't feel as good. Which is exactly the point.