1. Forum
  2. >
  3. Topic: Chinese
  4. >
  5. Consistency.



So I've come back to Chinese after I want to say a year? Been a while. While it was in beta, there was a lot of work to do, obviously, and it can't be easy to enter every conceivable possible correct way to phrase a translated sentence while developing it.


I'm not a newb to Chinese, long since having completed my education in it at the master's level. And I've found, just taking the placement test, that there are some problems that have to include "de" to be correct. Which of course is grammatical, but not necessarily as spoken or colloquial. Then there are problems where including it is counted as wrong, and the colloquial form is what was wanted.

Guys. Come on. There are a number of criticisms I've seen about having to basically do it without pinyin, and while it doesn't bother me personally, I can see how that's a problem for people who don't have a background in chinese already. But if you don't grade things to a consistent standard... I think that's a bigger issue. One thing is a steep learning curve, the other is not being sure what you're supposed to be learning in the first place. Oh, and then the "yao" vs "xuyao" thing. Did that get fixed?

It's a work in progress, and a lot better than it was in beta. And still needs work. At any rate, keep making it better, and I hope consistency will improve. I certainly do appreciate how much more of the language there is on duolingo than some languages here that basically are a phrase book. That's a big job, thanks for taking it on.

August 17, 2018


[deactivated user]

    Unfortunately, the improvements stopped several months ago.

    Duolingo relies on user reports to improve its courses. The users report questions with errors or missing answers. These reports get reviewed by course maintainers (almost always unpaid volunteers, although a few courses are maintained by paid staff), who make corrections or add new answers as needed.

    Something happened a while ago, when the maintainers for Chinese stopped answering reports. Maybe they had a falling out with Duolingo staff and quit, or maybe it was something else entirely. Nobody knows exactly what happened except the staff and the maintainers themselves, and they usually don't communicate these things to users.

    Edit: Whatever the problem is, it's not a lack of volunteers. I was given to understand that Duolingo gets hundreds of applications for every maintainer position.


    Assuming they're all volunteers, it's understandable. Like, it could be fun for a while, but after a certain point it's not like it'd be particularly rewarding. Not sure what value you can really claim from it besides just volunteer work.

    Something must have happened though, as you mentioned, cuz if you look at the course (https://incubator.duolingo.com/courses/zh-CN/en/status) there's fewer contributors than there even used to be.. there used to be 8 or 9, now there's 6. Granted the ones that left were probably only 1-2% by contributions made each.

    The whole contributor system is probably just a bit flawed, and its flaws are exasperated in a course like this. Like, each contributor has to be trusted to be the arbiter of both languages. Where, if you've gone through the entire course you'll know, there's a few areas where it's clear that whichever contributor made it has a questionable grasp of English. There's also other problems where the issue is caused just by words existing or not existing in different dialects (i.e. American vs British English, Chinese dialects in Taiwan, Beijing, Shanghai, etc.).

    The only really good way to fix that, besides having more obsessed/paid contributors, would be to just have a system that's more better crowdsourced. Like, instead of having a "report" black box you could have something more like an open wiki where people could discuss and vote on alternate translations and other changes.


    Yet, the dates the contributors deal with the reports are not as predictable as for the other different language teams. Unlike few language teams (such as Korean) who keep trees and reports up to date, Chinese-language team does not have the strong emphasis with structures of the course. Based on my reasoning and thoughts (I rely on logic more than guessing-and-checking), here are the possibilities why the tree is not well-structured:

    • Experience: Regardless of long-term experience with Chinese, it is possible that the contributors do not gain enough experience to construct the trees in Duolingo. Teaching Chinese in Duolingo is not the same as teaching in real life since in Duolingo you are dealing with learners across the world. Since you are dealing with short pool of learners in real life, you won't have a lot of pressure to manage how you teach them. It seems that Chinese contributors did not realize this difference before they started "hatching" the tree.
    • Characteristics: After reporting and also checking the pool of reports, I realized that learners are complaining mainly about missing alternatives and incorrect translations. Only some reports were dealt, and some problems were corrected. Yet, there are still more issues, pertaining to how the sentence is translated. From what I learned, not all contributors of different fields have same/similar type of characteristics and habits as others.

    On a positive side, Chinese tree improved a lot since it was released in beta. Even though the number of constructive/critical reports is still rising, they help us learn from other reporters. Whether the reports are true or false, they are the best resources for anyone at any level to read. The more you make or read comments, the better experience you will earn. What is seriously wrong with making the reports that are least likely to be taken if you have anything correct or interesting to say? At least, your worthy 2 cents in the report box can make a difference. ;)

    Learn Chinese in just 5 minutes a day. For free.