Monthly Archives: July 2017

This morning I noticed a feature change in the Duolingo app. When I made a mistake, the same question came up again immediately. In the past, the app would move on to the next question but would save up the wrongly answered questions and present them again at the end of the lesson, in the order they first appeared. I rather prefer that, to be honest. Getting it immediately after having been given the right answer seems just too easy.

I guess it is part of the process to make people feel good about using Duolingo so they don’t quit. I can certainly appreciate that, because once you give up, you learn nothing, and anything is better than nothing. But for me having to hold the correct answer in my working memory for a minute or four seemed to make it more of a learning experience and less simply a mechanical act of tapping it on the screen.

This was in the Android app. I tried doing some exercises on the iPad to see if it was the same there, but after five full 10-point exercises I still hadn’t had the chance to find out. Because you can’t go wrong with Apple. (No, but seriously I am starting to wonder if the IOS version is just plain going easy on users.) Anyway, I fell asleep in my chair at that point and woke up deciding that it didn’t matter whether it’s on this or that platform. I really think it is a bad idea no matter where you do it: If people can’t learn from a mistake to the point of getting it right after four minutes, they have probably run far ahead of themselves and need to slow down and repeat things. Besides, if you get it wrong the second time, you get the same question a third time. It has happened to me, but very rarely.  Seriously, if this is a problem, the solution is not to make it impossible to go wrong a second time, but to go back to the basics.

So say I, but I am not a highly trained expert. It may just be my smart privilege talking.

The Android app no longer uses 1-minute timed exercises as the only option for general practice. Evidently it was a test (some people never had it at all) and it seems to have failed the test. It was universally hated on the forums, to the point where people went back to earlier version of the app and installed it from unofficial sources rather than deal with the hassle. Hopefully they have disabled it for all users by now.

So this means I now get the standard 10-point exercise and have all the time in the world to finish it. The experiment came at a particularly bad time for me, since Japanese is more time-consuming than any of the other languages I have tried except possibly Hebrew. Like Hebrew, Japanese has its own script (except Japanese actually mixes 3 different scripts in normal text) so you first need to decode the writing system. Since you keep learning new kanji throughout the course, there will never be a time when you can just look at the text and know how to pronounce it, the way you mostly can with French after a few days.

Anyway, the app is better now, long may it last!

Also, I bought a subscription. It is as expensive as a full-fledged online game, so it’s not remotely worth it. But if I as a Norwegian doesn’t pay for it, who will? People in Guatemala or Vietnam? Since Duolingo in practice is nonprofit, at least with the current owners, the more subscribers the less they need to show ads and the more they can invest in expanding the services.