Hardly Any Translation from English into German
I've been having a real problem with the German course on Duolingo that has me wondering whether it's worth it to continue. In all lesson types on both the desktop and the Android app the German lessons are disproportionately comprised of a German sentence requiring translation into English. When it is a translation into German it is almost always a multiple choice (pick from 3 or 4) on the desktop and "pick the words" on Android. That is to say, it is very infrequent that I have to actually type out a German answer. I would say my anecdotal average is 1 time per lesson, as often there will be no typing lessons, and sometimes there is 2 or 3 (out of 20+) but usually there is exactly 1.
My issue is that I am not learning to produce German. I am learning to read German and translate it into my own language (English) but I am not learning to produce German. I'm getting more out of drilling vocabulary on Memrise than I am actually using Duolingo. Importantly, I'm not really getting any practice constructing proper, non-trivial, grammatically correct sentences.
To that end, using the German course on Duolingo doesn't seem worth it. It's not adding any value to my language learning process. This issue appears to be unique to the German course: I've tried both Dutch and Swedish recently and neither course had the same problem. They both averaged at least a quarter of the lesson typing Dutch or Swedish.
Is there any reason for this disproportionate bias against learning German (other than catering to school-children and building higher retention numbers)? Is there any way my data blip can make more of an impact other than not using the German course any more?
I have the same issue with Italian. I would recommend you try the reverse tree (where you learn English for German users). While I have not started the German reverse tree yet, I am doing the Italian reverse tree. Most of the sentences are in English and must be translated into Italian. An added bonus is that there are some different vocabulary words included. The only downsides are that it is a bit cumbersome to switch between the lessons and the comment sections are in the other language (this may not be much of a downside, depending on how much of the other language you understand). Hope this helps!
I was going to suggest the same thing; try the reverse!
I started doing that in Italian even though I haven't finished the Italian for English speaker tree yet because I felt the same way you do with your German. An added bonus is that most of the comments will be in the other language and it's interesting to read the 'problems' others have learning your language. That in and of itself may help you wrap your head around how their language works.
Give it a try!
I have noticed the same, but my current main goal is to understand German conversation and writing so it isn't really an issue. I intend to try and finish the Duolingo tree and then look at some of the other courses around.
However, how do you reverse the tree? Is it necessary to sign in as a different person?
Is it necessary to sign in as a different person?
However, how do you reverse the tree?
Go to the your profile settings page then to the Learning Language tab and choose the course you want in the drop-down menu (and save the changes).
you can also see here various methods to quickly switch between courses.
Update [2017/05/02]: here a script allowing an easy and quick Switch between courses.
Thanks for the reply. I had a look, but became nervous and worried that I would lose my settings. The methods to switch look a bit complex at first glance.
The first one may be a little complex to set up but after that it's the simplier to use.
The second one isn't that complex IMO: it's just typing (the URL address manually).
I, for one, COMPLETELY agree. I'm guessing it might have something to do with how they try to tweak things so that they get and keep more and more unique visitors. I wish there would be like a little slider to make it so that I could focus on more English to German and vice versa.
Agree wholeheartedly. I completely stopped using English as my base language for the same reason. The issue as been brought up many time by many users, but unfortunately, it's been rather silent from Duolingo on initiatives to improve this. So I suspect they are fairly set on favouring user retention over the effectiveness of the learning.
I totally agree. I also think that when any noun is taught it should appear with the gender.
... and the plural form. Everyone recommends learning vocabulary this way and then immediately fail to provide the noun words in this format. At the start you think its an overkill but unless you know this amongst many other things you cannot write correctly or begin to speak fluently. It continues to make using German a challenge
Has Duolingo recently gotten worse in this regard? I have been noticing similar issues with other languages. Translating from L2 (the language you are trying to learn) to L1 (the language you already know) is the easy part. The hard part is translating from L1 to L2 (or, even better, working entirely in L2). But in the practice I have been getting mostly L2 to L1. Using the reverse course is not a great solution, because the course should be organized around the grammar issues of L2, not L1.
It's almost as if the interests of Duolingo are not aligned with the interests of its users. Duolingo (at least according to its orginal goals) wants to train users to translate written texts from L2 to L1 (translation always works better in this direction because it is easier and the output will be much better). But users want to learn to function in L2.
But users want to learn to function in L2.
hard to be sure that it's what the majority of all the person having opened an account (= users) want the most and even less if it's what will work the best in average for users (independently from what, in average, users think is best for them).
The fact that there is more L2 to L1 than L1 to L2 results from A/B testings that have displayed better result (for this way of making things) on the metrics Duo is using. And the metrics include one that is "are the users coming back to learn the next day". So there is a competition between
- users that will not abandon after the first fail (some would say "users that really want to learn")
- users that need to be presented easy things and go slowly, and easily abandon otherwise
Note: Not saying I don't want more L1 to L2 (all the contrary) nor am I saying Duo is (or not) doing the right thing for the majority of users. Just explaining why it is how it is. ;)
Moreover, I imagine that the fact that the "motivated" users have a workaround (taking reverse course), Duolingo see even less reason to make things differently:
- group "1." can go to the reverse tree and keep leearning
- group "2." works fine with the "mostly L2 to L1".
What if there were an option in the settings where you could specify how much L1 to L2 versus L2 to L1 you want? It could also say something like "harder" and "easier" next to the options.
I'd like it too and the idea shows up regularly on forums. But Duo (still?) didn't implement it. Likely for the same kind of reasons, among which possibly:
- they don't want to
- it's somehow already doable: "easy" = take the tree and "hard" = take the reverse tree (Note: not working for courses for which Duo decided not creating a reverse tree, like SV, DA)
- they'd like to do it but they have a lot of other things more urgent (in their opinion).
This is exactly what I've been thinking actually. An option like this to change the balance for "power users" would be the ideal solution. Don't change it for everyone, don't even make it default, just give people the option. Oh well.
Oh, one other point, I just have to chime in as well with my obligatory criticism of their growth model. They seem to have completely lost the forest for the trees when it comes to the quality of language learning on offer. They might have a growing user base but how many of them are actually learning anything?
I think I still have a hang-up on using the reverse trees. I'm not here to learn English. I'm here to learn the myriad of other languages on here, many without a reverse tree.
Either way, I'd actually be fine if it was "mostly" L2 to L1. It's absolutely disproportionately nearly exclusively L2 to L1. That's the problem.
many without a reverse tree.
For the moment, we have to be patient as native speakers of other language than English are patient to have more courses for them. ;)
It's absolutely disproportionately nearly exclusively L2 to L1. That's the problem.
Duo evaluated it was not a problem for the majority of users: they evaluated (with their own criteria) it was beneficial to the vast majority.
So there is nothing that can be done there since they rely on their metrics: are they wrong to do so, maybe, but it doesn't change the fact that they do and thus it'll not change as long as they do.
the only way to have it changed isn't to ask for a change against their metrics but to ask for a change of metrics. Will they follow the "request"... not sure: it's asked for years...
Yeah, the evaluation is certainly that it's not a problem for most people. But my feeling is that the majority of people just like it as a game and aren't serious language learners.
I've felt like a guinea pig in Duolingo's experiment for years on here. It would help if they gave people some options or some clarification to their metrics and asked for feedback.
I'll also clarify, I'm not trying to change your mind or get anything done, just complain. I have no illusions about the likelihood of anything changing. But this topic is currently the most popular in my feed and on the German feed so hopefully people just think about it. That's all.
asked for feedback
IMO, not scalable given the number of users. That's why most companies with very very large user base wanting to "test" options do use A/B tests.
They are already struggling to deal with bug report, they would drown if they ask for feedback, so do it in an "invisible" way for users with continuously more or less 50 A/B tests running.
I'm not trying to change your mind
You wouldn't have too, I'd also like more "write in target language" exercises.
As a course contributor I'd like for them to be at least not almost totally discarded: I wouldn't feel to have lost time working on those exercises. As an user, I'd like it to be the majority of exercises.
It's just not like that.
I agree with your sentiments. The level of complexity in DL is not challenging enough to take a user beyond a basic level (I would say A2). I believe this is partly due to the fact that German is a difficult language because of its complexities. So it is either too much effort for the webmaster to provide the material for more advanced levels or too difficult for the learner and the fear on the part of DL (like so many other sources) is that it could put users off from learning. So they want to liked rather than admired for their rigour. To translate a German text into English, and vice versa, the latter of which is the true test of mastery, cannot be done in DL. Also the prompting by Duobot in Immersion does not really help a serious learner either as you think you make progress by scoring points when in fact you cannot produce the same independently. You have to do this in other forums.
If you really want to learn a language duolingo shouldn't be a stand alone learning tool. When you find other sources to learn from this problem won't feel so bad. Still duolingo will help to keep the pace and keep advancing.
Oh, Duolingo isn't the only thing I'll use, it's just the most convenient. Really I mostly use Duolingo for output. I have tons of input sources. I don't need another gamified version. I want to actually produce what I'm earning.
Perhaps you could try immersion... upload an english article and translate it into german.. people will fix your mistakes and help you to translate...
it should work as described here:
I have only tried it from german to english, but I suppose it should work the other way around as well.
I had not noticed what you point out. So I decided to collect some data.
This was from a mix of several sequential lessons, both new and "strengthening" exercises using the desktop/web (not mobile) version of DL. (A total of just over 100 exercises.)
German - English (typing): 57%
English - German (typing): 18%
listen type (in German): 14%
English - German (pick from list of translations): 6%
English - German (single word, pick article and type word): 5%
So while G-E (typing) is about 3 times as frequent as E-G (typing), I still get a fair amount of E-G exercises, especially if you add in the "single word" and "pick from a list" types of exercises.
Perhaps it has to do with where you are on the tree, and if you do the "strengthening" exercises to keep each lesson "gold". I'm about 2/3 or 3/4 down the tree, and I do a lot of strengthening/review rather than just powering on to new lessons.
All forms of exercises are needed in order to understand printed matter in German, understand spoken German, and ultimately express yourself in German.
I love German!!! Don't drop it!!! Try using other websites!!! Or work through it
I like it too. I wouldn't drop it completely from my learning map, I would just drop it on Duolingo.
I agree, as many other users have said. My reading and general understanding of German is pretty darn good because of here.
My other skills like writing and speaking? Nope, not so fantastic. The reverse tree is basically the same way, you write more English than you do German.
While I'm complaining here: If you do want to work on translating from L2 to L1, then the immersion feature is excellent for this. Unfortunately, most of the immersion articles (at least in the languages I have been working on) seem to be translations of things that were originally written in English. Translating these back to English again seems retarded. Some more interesting content in L2 would be nice.
Why are there so many here making excuses for DL? They are a product/service provider. They either deliver or they don't. If a lot of users are complaining about something, it's not a misunderstanding, it's a failure in service/product quality relative what users are asking for. And sketchy comments referring to survey results is just a smoke screen. Easy enough to focus surveys on users who have not yet reached a level where this problem becomes an issue.
Seems pretty straight forward. DL choses to focus on getting people started, and keeping them in for a short to medium period. Build user numbers, then use that to get more funding or sell. And right now, it is indeed a great tool for getting started with learning a language. However, there is clearly no commitment to helping users developing skills to any sort of advanced or even intermediary levels.
I'm very new to using DL (6 straight days) and this issue is already becoming a serious problem. Unless I very soon find ways to use DL for actually developing my skills, which with total certainly will require a lot more L1->L2 writing, I will soon stop using DL. Will see if there is more L1->L2 as I progress over the next 6-10 days, but if not, it will be a waste of time to continue and I will leave DL and switch to something else. No biggie for me. I will have had fun with it for a couple of weeks, free of charge, learning some very basic stuff that will soon be forgotten unless I engage in some other learning product.
If I don't see that development, which judging by these comments and what I have seen so far I will not, then I hope DL get the funding they must be seeking before their way of inflating user numbers implodes. I also hope that they use that funding for improving the product with regards to more advanced levels, ie L1->L2. Otherwise, this is just a switch-&-bait that will entertain for a short while, then disappear. That would be a shame and a waste, because it is a good start towards what could be a great and game-changing product.