My wife quit Duolingo the week before this announcement after years of watching Duolingo prioritize attention manipulation over learning. She had a nearly 6-year streak and was on the paid version at the time, but realized that it wasn't actually helping her learn any more: she'd at some point begun maintaining a streak just for the sake of maintaining a streak.
The best documentation for Duolingo's decline is this article from a few years ago [0]. It's a piece by Duolingo's CPO (who was a former Zynga employee) where he discusses at length how Duolingo started using streaks and other gamification techniques to optimize their numbers. He has a lot to say about manipulating users into spending more time with them, but in the entire piece he barely even gives a token nod to the supposed mission of the company to help people learn. The date he cites for the beginning of their efforts to optimize numbers pretty closely correlates to my sense for when my wife began to complain about Duolingo feeling more and more manipulative and less and less useful.
This past month they finally jumped the shark and she decided to quit after 6+ years. The subsequent announcement that they'd be using AI to churn out even more lackluster content gave us a good laugh but was hardly surprising: they'd given up on prioritizing learning a long while ago.
I got started on Duolingo back when it was still a "Help translate the world" app. I've always liked it for getting to dip my toes in a language and learn some basics whilst exploring the language myself through other methods, and I've shown my support of it by paying for Duolingo Super or whatever they're calling it for years on end whilst hopping on and off my language tracks.
But it's just so horrible now, constant gamification, attempts to pull me in with streaks and freezes and notifications and "did you know you can have us nag you even more"-breaks between the lessons I'm actually there for. It's gotten to the point where I'm just done because I've already paid for the service and i just want to be left alone to do the exercises, but they never let me get from one exercise to the next without having to go through at least two or three of those annoying "gamification and engagement" attempts.
Some (but not most or all) of Duolingo's social and gamification features/social nags/upsells/"reminders" default on but can be turned off in the settings. But yes it's out of control and a strong reaason to disable Auto-update on the Duolingo app to not constantly the ever-more-AI-driven-nags/upsells. DL is becoming its own antipattern in the quest for revenue $$$ growth at all costs, e.g. reducing the actual amount of language being learned, beyond a certain plateau. I've been saying that here for a couple of years:
When I returned to Duolingo recently -- I used to use it heavily but set it aside for 2 years -- I counted 14 gamification popups in a row after my first lesson in a new language.
14! The damned popups lasted longer than the lesson had!
I switched over to Busuu, which has blatantly copied some of Duolingo's mechanics but at least uses them with a modicum of restraint.
I disabled all possible notifications hoping I would only have the streak reminder, but no - it still abuses them with random crap. I then set an iPhone reminder for the streak, and completely disabled duolingo's notifications from the phone settings. Peace.
It still spams you after every lesson, but I often just kill the app when it does.
Quite a few ads also fail to load due to Lockdown mode or my pihole (also when away from home, due to the vpn I always keep).
I may just be their worst customer, having never given them a cent or even clicked an ad (and often not even impressions). On the other hand a bunch of people use it because of me and follow me due to having a long streak, so maybe I'm still worth keeping around.
Learning a new language to any degree of proficiency requires motivation. It's easy to start and hard to continue if you're not willing to put in the effort.
There's a valid argument to be made that gamification helps to provide that motivation, but the argument doesn't hold up if the users aren't actually becoming proficient by using the app.
In other words, gamification isn't inherently bad, but their motivations don't appear to be good.
So I agree they go over the top with it, _but_ I reached fluency in Spanish in about 2.5 years and Duolingo was an indispensable part of it.
> if the users aren't actually becoming proficient by using the app
Learning a language to fluency requires real commitment, and I’d say an app could never possibly do it on its own. One of the most key things Duolingo gave me was consistency and a lack of an excuse to constantly practice and learn. But you also have to (and I did) use the language daily, watch content in the target language, travel and speak with locals in the language, etc. I’m not sure where Duolingo ever claimed that it alone was enough to actually reach proficiency or fluency.
Duolingo’s gamification and streaks and leaderboards gave me a reason to put a lot of effort into learning the language, and I don’t know where I’d be without it. There’s a lot of things about Duolingo I don’t love but I’m incredibly grateful that it exists.
Agreed. Duolingo started out on the right foot: they had gamification, but not too much of it, and they clearly cared about helping you learn. For a long time it was the most highly recommended app for learning new languages and that wasn't just naivete, it actually did work.
That's changed gradually over the last few years as they switched from using gamification in pursuit of learning to using a veneer of learning as a pitch to get people to try their game.
I sort of wonder whether they realized gamification works for certain kinds of tasks, but not for others, and then decided to design their language learning app for gamification, rather than designing a gamification system to support language learning. In other words, I don't think Duolingo's system can really make you fluent in a language, but what it seems to excel in is making you use Duolingo every day. In other other words, you always hear people talking about how long their streak is ("500 days!") rather than how well they speak the language.
It requires habit more than motivation. I was bored during covid and started Japanese and four years later im still keeping it going. I lost my motivation multiple times during those years (because good god what an ridiculous language coming from a European one), but my habit kept me going until i found my motivation again.
internal motivation means that someones acts without external stimuli, their drive comes from within, its internal. External motivation means that an external stimuli is used to make someone act. I.e. a monetary reward, or validation etc. When someone is internally motivated, they can have a stable state. When external motivation is introduced, it can replace the internal motivation and will. Now what happens when you then lose the external motivation, the external stimuli again? The internal motivation is gone and this means all motivation is gone, the act stops.
On top of this, some people say motivation is cheap, discipline is what matters.
I disagree with this in principle. Gamification is something we should be very wary of because it is inherently bad. It reduces what you care about in an activity to points and a progress bar.
Instead of sticking with language learning because you have some intrinsic reason to want to learn it (or even a external one such as wanting a new job) you're substituting that with whatever Duolingo puts for their gamification. To the degree you engage with and are motivated by the gamification you are substituting your own metrics of success and progress for points and streaks.
And soon enough we end up here, where Duolingo has gamified their internal numbers and in doing so gamified your "learning".
I caught a random podcast with an early Duolingo employee who said all the same things: Much bragging about how they gamified their app to juice user engagement and growth, not even a feigned mention of optimizing for learning.
By now my friends who use Duolingo all know it’s a game, not a real learning experience. I think they got lucky and filled a void in the market for things people think they want (learning a new language) while avoiding the parts they dislike (the effort of learning).
It got recommended by default for years when people asked for an easy way to learn a language, but they leaned hard into the path of gamification instead of trying to improve the learning experience for those who wanted to learn.
When Duolingo added that viral post on Growth hacking, it caused quite a stir about the push-notifications and gamification tactics they use. Ultimately, we decided it wasn't worth it for Coursera to veer into edu-tainment.
However, it is interesting to watch how much gamification works in adding and retaining users. In 2023, Duolingo's marketcap was 5x of Coursera. Now at similar revenue it is 20x of Coursera.
As a user, I think Duolingo is over-gamified (stopped using it) but Coursera is severely on the other spectrum where it comes off as too bland/boring to keep up the motivation. I am sure there's a happy medium to be found between reminding users to engage in something hard while doing right by learners.
> In 2023, Duolingo's marketcap was 5x of Coursera
Maybe this is because Coursera requires a lot more effort? My pessimistic view is that most people do not like learning, let alone learning in their leisure time with their own bucks. On the other hand, Duolingo gives more people the false impression that it's easy to learn a new language. And check out their math and chess program. They are really really easy, like pre-school level easy. Naturally, more people would be using Duolingo.
The leverage from revenue to market cap can mean a few different things. Duolingo may be overvalued, coursera undervalued, or both. In any case, the impact of gamification on revenue would be a better indicator of its direct efficiency. The rest is just creative charts on a slide.
Duolingo is really only useful at the A1/A2 levels anyway. Once you reach B1, you're pretty much past the point where the vocab and grammar basics from Duolingo is useful and you need to move on to other activities (watching TV in your target language, having conversations with native speakers, reading books in your target language, etc).
Early on, the Duolingo stated goals was to teach language to the point where it's learners could ultimately start translating documents. They were going to sell cheap mechanical turk style translation services. (Think captcha style translation)
Unfortunately as they got popular automated translation services got good enough that nobody was going to pay for a slightly better and slower translation enmass.
Once that happened, that's when it seemed like they dumped their goals of teaching language and instead focused on dark pattern money extraction.
To be a bit cynical about it: the typical DuoLingo player has probably been misled to some extent about its effectiveness, yes, but also many of them don't particularly want to learn a new language. I suspect that they're happy to be able to play a popular mobile game that everyone else is also playing without the stigma of being a "Candy Crush addict" and "timewaster". "I'm learning a language!" is the welcome figleaf. https://youtu.be/F3SzNuEGmwQ?t=243
I disagree. In Spanish, learning the subjunctive is essential and that’s part of B2, and I think Duolingo did a good job of teaching it. If you can’t understand “Que te vaya bien” even completing a purchase at a store would be a bit difficult.
> Once you reach B1, you're pretty much past the point where the vocab and grammar basics from Duolingo is useful
It drives me nuts that Duolingo's Japanese course does not explain grammar nor does it introduce new grammars fast enough. It's super boring to see です and ます most of the time, with occasional new grammar points thrown in. It's also strange that Duolingo introduces honorifics without context. This is super confusing. Who in the world would decipher a long string of characters out of a few really bland sentences?
Which would be perfectly fine. A1/A2 require plenty of time to master. I know the internet is filled with people going “i learned A2 in one week” but that doesnt mean that its really internalized.
It seems like with sufficient funds, Khan Academy could offer this experience (language learning) without the enshittification Duolingo demonstrated. Think how Evernote faded away, but for different reasons.
My girlfriend has been "learning" a language on Duolingo for about 5-6 years now, but rarely engages with her target language outside of Duolingo and some of her music library. She's been at roughly beginner level the entire time when, with proper language immersion and practice, she should realistically have a large vocab and be able to engage in casual conversation without looking up stuff. This is not the case.
I've just accepted for some time, to her chagrin, that she's effectively playing a game that just so happens to be language themed.
That's a feature, though. If she's too disinterested in making other moves like reading books or the news, then at least she still has that base Duolingo momentum that might make the move possible in the future.
People always assume the alternative to Duolingo is that everyone will start a habit of reading BBC Mundo in Spanish or something, and it's obviously not true for many if not most people. And that's fine, some people are only going to scoot by with a dilettante level of interest until they take a real plunge.
5 minutes per day for a year or two is about equivalent to ~2-3 weeks of traveling to your country of choice and just trying to talk with people on purpose. It’s probably easier to do in Italy than in Finland, but ultimately nothing beats just being there. Duolingo might just be enough to give you an okish on-ramp to that experience.
I've got a 3 year streak and gamification was obvious to me on day 2. Some of their feature flag experiments are very in your face, too.
Still, 3 minutes per day is just about my tempo. I don't care about literally anything in there except the learning part and consistently doing only one lesson per day makes them very nice and polite most of the time - I feel like I'm in the 'beg-these-for-money' instead of 'milk-them-dry' cohort. (Or maybe I'm in the permanent 'lets-be-nice-for-them' long running experiment?)
It doesn’t have to be that way; learning a language is a long process. I took about 3 years to reach real practical fluency, and I still have to say “what?” more often than I’d like and I still need to learn a ton of more advanced vocabulary. Duolingo unfortunately doesn’t offer any C1 content so I’m stuck using other methods.
Mindless optimization of basic "attention grab" metric is why the whole internet feels like a slots machine. Be it reddit, Facebook, YouTube, any google result
Thankfully this won't happen with LLMs, as compute is too expensive so execs can't just take an easy way out of optimizing for number of questions asked
A real question: did she actually learn a language? 6 years should be enough to be fluent at any language. My opinion is Duolingo just doesn’t work and never will. People are fooled by the gamification but it’s a time wasting app/game that gives an illusion of productivity. Like Minecraft with words.
Exactly, I did the free Language Transfer audio courses in Spanish and French and actually learnt some real useful basics of the language (they're very much introductory courses but very good).
Previously I'd spend far more time in Duolingo trying to learn and hadn't really learnt any useful language skills at all. I can see it being helpful for drilling some vocab if you were learning elsewhere, but it just doesn't work to actually learn a language.
> how Duolingo started using streaks and other gamification techniques to optimize their numbers
These two tactics per se are alright, right? If anything, I'd appreciate that Duolingo tries to keep me engaged. Besides, the more one spends time on learning language, the faster they learn.
The issue with Duolingo is not about gamification, but that translation is ineffective and boring, no matter how much gamification there is. Personally I find that the most effective way to learn a new language is starting with Comprehensible Input and then moving on with tons of output. Take Spanish for example, Easy Spanish, Dreaming in Spanish, Español Sí!, Extra, and Destinos offers lots of fun input for beginners. Paco Ardit's graded readers are great too.
Another problem with Duolingo is that it does not help listening comprehension at all. It turns out that we can only pick up sounds in context with tons of repetitions and combinations in consecutive sentences - a feature that is exactly what Duolingo misses. Yes, it has introduced listening and stories, but the amount of them is too little to be useful. Another lesson is that reading does not help improving listening much. When we read, we see individual words and phrases easily, while it's really hard to pick up individual words when listening. I didn't understand the difference and spent a lot more time reading than listening. As a result, my reading was at the level C1 yet I could only understand slow Spanish at the level of A2.
Right. Something that businesses don't appreciate enough is that while an unpopular decision may sound small enough that they shouldn't lose customers over it, not all of them were happy to start with.
The timing of when I finally quit Twitter was when they shut down third-party clients, but that was after I was already half checked out because it had been in decline for a few years already (predating the change in ownership.)
(As an aside I hate that videos are now "source material" for a discussion... it feels somewhat lame).
> wife began to complain about Duolingo feeling more and more manipulative and less and less useful.
I think this is a great bit of insight into what a lot of what the web has become! If they had been more manipulative and stepped up the quality and utility of the product would that have been acceptable to remain competitive vs something like tiktok?
> she'd at some point begun maintaining a streak just for the sake of maintaining a streak
A friend of mine said the exact same thing. And then a YouTube creator I follow recently made a video where he said the exact same thing too about cancelling Duolingo because he had become more addicted to maintaining the streak than learning.
A streak where you actually learn every day makes sense to me. Missing a day and then paying money to maintain your streak doesn't. I know I missed a day, and if the streak isn't for me, then who is it for?
Really seems like a lot of bad people came out of that company. It's surprisingly that they're widely accepted in the industry given how terrible that company's culture seems to have been.
This isn't my field, but I can imagine it is hard to optimize for learning only since the reward signal is clearly user engagement (meaning subscription revenue). Finding a reward signal that does both, help people learn -and- make money is hard. I am a Duolingo user and I definitely notice the gamification but I really don't know how it would be done better since that gets people engaged in an activity that is associated with learning. This is their whole job to find these signals, but honestly, what is it? What signal would you put in place that would keep users AND actually teach them something?
>… it is hard to optimize for learning only since the reward signal is clearly user engagement (meaning subscription revenue).
Schools can give reward signals for demonstrating subject mastery, or for tuition payment and attendance. It seems like Duolingo gamified the latter instead of the former.
Why education should be done for maximal profit? Oh the poor CEO, he just wants to bring more value to shareholders! Clearly folks are fed up with eventual inevitable result.
Things like education or healthcare shouldnt be privatized, since that always eventually ends up as profit-first game. The product suffers since milking is obvious, and quality of service is at best secondary concern.
> My wife quit Duolingo the week before this announcement after years of watching Duolingo prioritize attention manipulation over learning. She had a nearly 6-year streak and was on the paid version at the time, but realized that it wasn't actually helping her learn any more: she'd at some point begun maintaining a streak just for the sake of maintaining a streak.
I found the same thing with one of the meditation apps. I was just maintaining the streak, but not getting anything from it, after about a year. I can't imagine doing that for 6 years, so hats off to your wife.
When you think about it, if Duolingo does a good job of teaching language, people will stop using it. It’s the same problem with Tinder: people who stay together delete the app.
My Duolingo streak is 37 days and I just jumped on to do a lesson and retain my position on the leaderboard. I feel like the app itself is right on the cusp of being a valuable learning tool compared to being a silly game. I am okay with the idea of paying for “just okay” teaching if it helps me stay motivated and interested in the content. That may change in the future, I guess we’ll see!
What you're missing is that it's right on the cusp of being a learning tool instead of a silly game because it has regressed below the cusp. It was a useful learning tool, and its trajectory has been strictly downward for years now.
If it's working for you, don't let the negativity get you down! A year of daily Duolingo got me far enough along that I can now generally follow YouTube videos, news articles, and Reddit threads in the language I've been learning, with the occasional dip into Google Translate for unfamiliar words or phrases. The feeling of being able to just listen or read, and not have to consciously translate, felt like a door opening up in the world - a powerful motivation to keep working at it. Duolingo may not work for everyone, but in my experience there really has been value in it.
I found that with other study options, I learn faster and get tired less quickly (since they are less “puzzle-y” and more just language learning plain).
It's a little bare bones for someone young, but you could try Anki. It's a generic spaced-repetition app, so you would need to grab a deck of flashcards from the AnkiHub community for your language of choice.
For apps, I use Clozemaster and Babbel. Unfortunately Babbel is starting to feel like it's also chasing gamification, but it does have sensible content.
For podcasts and YouTube content, I follow EasyGerman and Coffee Break German, and both are part of larger brands for other languages:
Not knowing what language she’s learning, it’s a bit tough to say. Many have an app with lots of reading material with audio and assistance tracking learned words, tap to dictionary lookup, etc. It’s a pretty good category and a lot of kids enjoy them.
I 100% get this, but I will say that the gamification has meant I've stuck with it for a while. I don't know if it's out there, but I'd love something with better teaching methods, but just enough gamification to keep me going.
Duolingo, Facebook, Google, yada, yada. I'm not sure why, but it seems like when a vc track co. becomes a unicorn they game their own system into the ground almost every time. (no pun intended)
> barely even gives a token nod to the supposed mission of the company to help people learn
This was the impression I got from the app the first time I tried it, which had to be some time before 10 years ago (not that I'd suggest nobody gets any value from it, I assume mileage varies). It just seemed to reduce an inherently arduous and deliberate effort into something that was primarily easy and gratifying. There's a very lucrative market in convincing people they've learnt rather than entertained, whether it's through easily digestible YouTube videos (a trap I've fallen into) or apps that intensely use gamification elements.
There's a difference between the concept of delivering better methods of learning or easier access to good information, and delivery mechanisms that try to make it as fun as scrolling Instagram, in my mind anyway. Coursera actually has some really solid learning paths, and free university lectures are invaluable, but even those aren't effective if you don't deliberately allocate significant time and mental energy into grinding through the stuff you don't understand, and this seems just as true for languages.
So it just always seems like a false promise to me, at least beyond the premise of making it more approachable at the outset, and I guess it doesn't surprise me that things seem to be further going in a lame direction.
For me, the problem with Duolingo has always been that the content is just too lowest-common-denominator, and this will just bring it down even lower.
I switched a while ago to Seedlang (https://www.seedlang.com/), and while it only supports French, German, and Spanish, I can at least say that the German course is everything I actually wanted from Duolingo.
Every exercise involves a real video of a real German speaker speaking in German. You get to hear them at the same time as you see their face, which is not something you'd think is a big deal, but absolutely does make a big difference.
When it's your turn to say a phrase, it records your voice and plays it back to you, rather than use some shitty model to try and guess if you spoke correctly. By listening to your own voice you can clearly hear when you're getting things right versus when you're getting things wrong. Early on, German speakers would often comment on how my accent was quite good for my level, and I think this is big part of that.
IMO Duolingo's attempt to try and scale to every language as fast as possible just makes it a worse product than something 'artisanal' like Seedlang (though of course, if there's no artisanal resources, then Duolingo might have some value to offer)
> For me, the problem with Duolingo has always been that the content is just too lowest-common-denominator, and this will just bring it down even lower.
Interestingly, this is indirectly mentioned in the Linkedin post: "I've always encouraged our team to embrace new technology (that’s why we originally built for mobile instead of desktop)"
"Mobile-first" anything has always been a race to the bottom for everything: attention spans, information density and nuance, target demographic. Not just with Duolingo , but also with:
- Investing (Robinhood leaning into meme stocks and "gamification")
- Gaming (Angry Birds going from $3 lifetime purchase to a pay-to-win, micro-transaction hellscape)
- And of course, the first casualty, human communication (280 characters instead of an essay or open letter.
I studied German for 3 years in university, dabbled in German duo-lingo, and completed all German courses on Memrise.
I don't see how one can learn German fluently using Duolingo (or even Memrise, which I think is much better). It's great for vocabulary, but I think understanding the grammar requires understanding the theory which I didn't see when I used these applications.
Agreed. Even Seedlang was of limited use to me past a certain point, I just think it did a much better job at the same niche as Duolingo.
For me as someone who has never taken actual German courses, the biggest thing that contributed to my fluency was just listening to podcasts in German non-stop. Didn't matter if I wasn't understanding anything for months and months at the start.
I think the listening played a huge role in familiarizing my brain with wide swathes of the language. It made it so that when I learned other things later on, instead of being actually 'new', it was things I recognized and already had a sort of 'feel' for by association, even if I didn't know what it actually meant.
It was really cool watching as I went through a bit of a 'phase-change' at one point where one week I felt like I wasn't understanding more than few words per sentence and not able to actually follow conversations without looking stuff up, and then the next week it suddenly 'melted' and I was able to bridge the meaning between words and was actually understanding and following entire conversations.
My German still isn't perfect, especially my grammar and I probably should take some courses for that though. But I am at least fluent which is great.
I speak three languages and I'm learning a fourth. Don't study grammar, ever, it's a waste of time. Grammar rules always fall into one of two categories: the ones that are so obvious that you would have learned them after two examples anyway or the ones that are too vague and complicated to be useful.
For an example of the latter look up people making flowcharts for the subjunctive or for the ga/wa distinction.
Or, for that matter, find me the place in an english grammar that explains why you get "on" a train, but "in" a car.
Humans don't learn the grammar by "understanding the theory". Humans learn the grammar by using the language repeatedly.
But a book on theory can be mass produced and sold to everyone who wants to learn a language. Can't bottle and mass produce an actual experience of using the language for years. So theory it is.
I used Duolingo maybe ten years ago to get myself up to approx A1, mid-A2 level German. Back then every new piece of grammar actually had an explanatory page that you could study before jumping in to the quiz games. As the enshittification began, they made these harder to find so that instead encountering a new linguistic concept in the quiz felt like a cruel guessing game.
For beginners to German these days, I heartily recommend the free “Nicos Weg” course from DW that goes up to B1 at least. Also has, unusually for language classes, a cast of likeable characters played by reasonably good actors carrying a consistent, building storyline throughout the lessons.
I did every German module on Duolingo in a 5-month preparation for moving to Germany for my job, and I got the coveted Golden Owl to prove my proficiency...
... only for me to get to Germany and realise very early on that I would need to do a basic A1 language course.
The app was gibberish; the pronunciations were wrong, the genders were misleading, and the daily interactions they tried to drill in me were far from useful.
The overinflated proficiency instilled in me by the app, made me genuinely believe I could interact easily with a German - a delusion I was quickly and painfully made aware of, much to my chagrin.
I like that Duolingo gives me some kind of curriculum and guides me through new concepts so I'm exposed to new words and study them for a few days at a time. But you have to be intentional with your learning to actually learn. Learning a language is hard, and Duolingo knows people will stop using their app if they challenge people too much and the app becomes a place to feel bad about how little Spanish you know. So their lessons are designed to be passable and not frustrating rather than a method of learning.
Some techniques I use is not looking at the words when they're being read out to practice my listening (though sometimes the TTS voices make things unnecessarily difficult to understand), and I also try not to look at the word bank before trying to translate a sentence in my head first.
My main wish from Duolingo is some kind of lesson I could go into that just grabs questions from old lessons with words/phrases you haven't done in a while. It's a little too easy to get into the swing of a unit where the words are fresh in your brain's cache, but having them pulled out from cold storage would makes sure you've actually got them locked into your memory.
Also they should have a setting to disable word banks so you're forced to type everything.
Seedlang seems cool though, I'm gonna give it a download later.
Seedlang also has a curriculum design (and one that I think makes more sense).
> My main wish from Duolingo is some kind of lesson I could go into that just grabs questions from old lessons with words/phrases you haven't done in a while. It's a little too easy to get into the swing of a unit where the words are fresh in your brain's cache, but having them pulled out of nowhere would makes sure you've actually got them locked into your memory
Seedlang does this too. There's a gigantic library of all the exercises and you can go through them and put them in your review lists. Each time you review an exercise, you rate the exercise as 'hard' or 'easy', and depening on the rating, that exercise will then show up more or less often in the future. Eventually if the interval gets to be a year long, it'll give you the option to retire an exercise.
Each time you do a lesson, it'll list all the exercises from that lesson and you can choose which ones you want added to your review queue. It's really nice. Lots of control over your own spaced-reptition needs.
> Also they should have a setting to disable word banks so you're forced to type everything.
> My main wish from Duolingo is some kind of lesson I could go into that just grabs questions from old lessons with words/phrases you haven't done in a while.
Ah, Duolingo. What lengths people go just to avoid reading a book and talking to people. All this tech and I wonder how many people are now fluent in multiple languages compared to say a few decades ago.
This is not academic. All signs are saying we are heading in the wrong direction[1] and more tech ain't going to solve shit. Literacy rates and numerical abilities are going down the drain faster than you can say "Claude". I suggest we really, really get our acts together and stop trusting tech to solve our non-tech problems.[2]
“It is actually hard to imagine that every third person you meet on the street has difficulties reading even simple things.” [3]
Duolingo is for getting to an A2 level. It’s hard for most people to get much value from trying to read a book or talk with someone unless they’re at least at A2 level
That's bad news for the millions that had to do just that and learned fine in fact better. I'm not saying you should read the Upanishads in the original, I'm saying a proper intro textbook is fine and some teacher-figure ("human") to actually use your body and read theirs.
Heck, Latin even has a Latin-only method: LLPSI. "Roma in Italia est", you'll figure it out quickly enough.
> What lengths people go just to avoid reading a book and talking to people.
odd take, do you have the same negative sentiment for the entire field of language instruction? introductory German class at a local college? 7th grade French ? things like that?
I don't think it is an odd take. I live in Switzerland, which has 4 national languages (Swiss German is spoken in dialect form that varies between towns, while Romansh has well-defined idioms with distinct spelling, although the 5 or so idioms are mutually intelligible). I speak French, passable German and one of the Romanshes, and I'm a native of none of them. Between French and Romansh I can more or less read Italian, although I can't understand it when spoken.
The same thing that has worked for me as a method for learning languages has always been the same. Get books, particularly short stories or children's stories aimed at A2/B1 level, and read them. Practice grammar. Get a pen and paper and learn vocab by repeatedly writing it down. Boring but effective. And of course practice listening and talking, which means either having native friends, doing a course, using audio materials from somewhere, etc. Courses with actual humans make learning go faster (in the case of Romansh, it would have been impossible without the course).
I don't find duolingo to be effective at all, as others mention beyond the A1/A2 level. I'd be a bit more skeptical and say even A2 you need to expand your horizons.
Side note: the linked articles are really disturbing, particularly in regards to the growing numbers of illiteracy worldwide and - absolutely astonishing for me - the extraordinarily high number of education/job requirements mismatch in Western countries (first link and then directly to OECD report). This suggests that we're doing education wrong with the US leading at 25%.
> authoritarian rule, often involving the fusion of state and corporate tech power, where technology is seen as the driving force of the regime and used to consolidate control, suppress dissent, and erode public trust.
I’m trying to connect the dots between the above and Duolingo.
I stopped using duolingo regularly about a month ago. It's wonderful that Luis von Ahn says in interviews that he tries to prevent teams from cluttering the app, but it seems like he lost the battle. You can get 10+ pop-ups after a lesson. The friend feed is cluttered with meaningless achievements. The web app is tolerable, but the phone experience is miserable. But if you're behind a computer and keyboard, there are much more effective ways to learn. Busuu is a much warmer product on either device, with videos of native-language speakers to help with listening.
Duolingo has scaling and distribution. It makes no sense to scrimp for pennies on a product (e.g. English learning Spanish) that has millions of daily users. The AI radio lessons feel alienating and demoralizing compared to voice-acted stories, and the quality control is much worse.
A CEO has the power to do anything, but employees have the power to collectively, quietly sandbag if they don't like the leadership. I think the AI effort led to a broad disillusionment, causing an unwillingness to put extra effort into their work. Across the company, everyone starts to take the path of least resistance. The CEO senses his influence waning and becomes more accommodating to avoid further morale death spiral. So situations can arise where a CEO would like a cleaner product (no one likes to ship garbage), but has lost the political capital to make it happen.
This might be a nice time for me to plug the FOSS software I've been quietly building over the last ~3 years for English speaking learners of Finnish specifically. I've recently collected them all onto a little landing page at https://finbug.xyz/ .
I run into other immigrants to this country semi-regularly now who say they've used at least one of these tools, most commonly the frequency deck or reverse-conjugator/decliner. It's been a surprisingly fruitful way to make professional connections here.
Their comments cemented to me that they have no long term value. If the ceo of Duolingo thinks AI will teach me a language then I’ll use a low cost LLM to get there without Duolingo.
I cannot imagine being the CEO of a software company and proudly proclaiming that some other software — which we do not make and can easily be had very cheap or for free— is better fit for purpose than what we’ve been able to accomplish in over a decade with almost a thousand employees, and expecting that to impress investors or users or both.
Like that would be like the Chipotle CEO proudly announcing that they’re firing their workers because they’re getting all of their ingredients from Taco Bell now due to “Taco Bell’s system being so much easier to operate” and “Taco Bell is so cheap and they have so many locations”
Chipotle already doesn't manufacture the vast majority of their ingredients. Their business consists of delivering supplier-originated ingredients to a standardized storefront in your city and consistently assembling them into a quality burrito, with small amounts of prep work for things which can't be consistently sourced locally and in-store cooking for things which must be prepared fresh. I don't know where people get this idea that packaging and assembly are trivial tasks you can't build a business around.
I do think AI could be a better language tutor than an average language teacher, but I don't think Duolingo's approach is very effective.
The ideal AI-powered tutor would work more closely to a private language tutor. It would speak with you and gradually integrating language concepts into the conversation. When you make mistakes, it could correct you on the spot and keep track of where your strengths and weaknesses are.
You should look at HelloTalk. It's real people communicating in each others' language. Unbelievably great service, almost as good as sitting down to a coffee with someone.
Actually, in some ways better. HelloTalk has a "correct the other person's sentence" feature that shows an inline diff, yet it's simple enough for people who have never heard the term diff.
Correction is not always desirable. The goal in learning a language is rarely to be grammatically "correct" in the language, but rather to communicate. And communication doesn't need perfect grammar.
When I was working as language teacher, I was tasked specifically with teaching speaking. I would often use information gap activities. These are activities where two or more parties have pieces of information but need to obtain pieces from others in order to complete the task. Sometimes, these would us language forms (re: sentence structures), but most of the time they were free flow activities. It didn't matter how "correct" the language was so long as the idea was communicated.
To think about it another, how often do we make mistakes when speaking? Writing? And yet, we still managed to communicate just fine.
That's not to say there shouldn't be any focus on form, but simply that it's not nearly as important as many tend to think when it comes to language learning.
Exactly, you'd think a targeted, application-specific, purpose built tool would be what a vendor would gravitate to, not probabilistic, non deterministic hype, hot off a shelf. I really wish we could have overlords with at least some technical knowledge.
I've noticed that a strong thread of the hacker community, including HN and the guys who wrote GNU and linux, are extremely cheap. Like, you'd rather write a product clone yourself than pay $15/month for the official product. Why is this? What's with the inherent stinginess?
If an LLM can teach me a language, why wouldn't I go straight to the source and use GPT or Claude and customize it to my exact needs.
I feel like so many AI products these days won't be around a few years from now once more people find out that all their doing is providing a slightly different UI to what you can get directly from OAI, Anthropic, Google for cheaper and better and more tailored to you.
I think the opposite is true: I'm deliberately choosing the usually more expensive DIY route even though I'm aware of cheaper commercial offerings.
If you hate ads, deception, and dark patterns as much as I do, then most software has a negative value. But I'll only pay for it if the value exceeds the price. That means there's a rather large (and probably growing) share of the software market that I'll just never pay even a penny for, because in my opinion, it has negative value. But it's not because I don't want to spend the money. It's because I want those companies that are peddling ad-laden bloatware to be a financial failure so that the market will intervene and offer better alternatives.
Lol. Linus didn't pour decades of his life into Linux to save a buck. Neither did Stallman with GNU. Even if the alternatives at the time only cost $15, (hahahahahaha) free alternatives would still have been well worth making.
Software freedom has never been about money, and always been about preserving public access to a software commons. You might believe your local bookstore is a better source of knowledge than your local library, but that's only because you don't really know what you're missing.
If you enjoy building things why not do it yourself while also saving money?
My dad likes doing oil changes himself which I'd never do, but it doesn't occur to me to insult and question why a person has different values than I do.
If you're going to drag Linus and Stallman into this, at least read up on your history. A commercial Unix or Lisp Machine was quite a bit more expensive than $15/month. They both got shafted by companies that couldn't care less about anything beyond profit. It's part of the ugly side of capitalism, fuck people and fuck the world.
I'm at a fairly high level in learning spanish, which mostly came from actually speaking the language and just studying the dictionary etc... Duolingo was absolutely useless - and that was 6 years ago. I can only imagine how terrible it is nowadays.
Anyway, Im responding to you because i find ChatGPT to be a FANTASTIC tutor. Like I was absolutely blown away. It can do all the translation stuff, but also answer questions about different verb tenses, conjugations, syntax etc... I'm sure that an extremely good spanish teacher would be better, but I think ChatGPT is probably better than most. And it is free.
But as you yourself admit, you were at a high level already, so you know how to ask the LLM good questions. If you were a beginner with zero knowledge, you wouldn't have that advantage, so the LLM's ability to help you would be far less. How many ESL learners are asking ChatGPT when to use the past participle or present perfect tense?
When I am learning something new in development, the LLM is immensely useful since it has unlimited patience, and I can zero it in on exactly the level of complexity or understanding that I need.
I was happily paying for Duolingo Super, despite being unconfident in its pedagogy, until they announced they were replacing their human curriculum writers.
Dropped it instantly. I get the bargain-basement cost-cutting appeal from a (bad) CEO's perspective, but if I'm paying actual money for a service, I want said money going to humans.
Not to mention, at that point they are essentially just serving as the middle man between you and a LLM. Instead of paying for Duolingo to ask an LLM to generate a bunch of Spanish phrases for you, you can pay for the LLM and do that yourself plus a lot more. I’m not sure Duolingo has a solid understanding of why they exist as a business.
The value proposition gets even more dicey if the target language is one with a sizable online learning community. Those are churning out more tools and instructional content every day, most of which are made freely available, and this content has the benefit of not only being made by humans but also by people who are passionate about the language who care about the small details. These communities are much quicker to embrace experimental learning methods, too, and so generally the stuff that doesn’t work gets filtered out in short order.
And as others in these comments note, if one wants to use an LLM as a language tutor, you can do that for less than what Duolingo is charging while also getting the benefit of being able to tune it to your exact needs instead of being stuck with whatever Duolingo decided is best.
If an LLM is what's going to teach me a new language, why would I ever pay some middleman $100-200 a year for an app wrapper? This guy doesn't seem to realize that embracing AI-first doesn't just put his employees on the chopping block, it actually suggests his whole company is unnecessary.
This is what I don't think the "AI-first" business crowd understands -- in many cases, the moment you admit the humans in your organization can be wholesale replaced by AI, that's a sign it's possible your whole ass business case could be unnecessary LLM middleware.
Readit News
The best documentation for Duolingo's decline is this article from a few years ago [0]. It's a piece by Duolingo's CPO (who was a former Zynga employee) where he discusses at length how Duolingo started using streaks and other gamification techniques to optimize their numbers. He has a lot to say about manipulating users into spending more time with them, but in the entire piece he barely even gives a token nod to the supposed mission of the company to help people learn. The date he cites for the beginning of their efforts to optimize numbers pretty closely correlates to my sense for when my wife began to complain about Duolingo feeling more and more manipulative and less and less useful.
This past month they finally jumped the shark and she decided to quit after 6+ years. The subsequent announcement that they'd be using AI to churn out even more lackluster content gave us a good laugh but was hardly surprising: they'd given up on prioritizing learning a long while ago.
[0] https://news.ycombinator.com/item?id=34977435
But it's just so horrible now, constant gamification, attempts to pull me in with streaks and freezes and notifications and "did you know you can have us nag you even more"-breaks between the lessons I'm actually there for. It's gotten to the point where I'm just done because I've already paid for the service and i just want to be left alone to do the exercises, but they never let me get from one exercise to the next without having to go through at least two or three of those annoying "gamification and engagement" attempts.
https://news.ycombinator.com/item?id=35287456
https://news.ycombinator.com/item?id=35297240
https://news.ycombinator.com/item?id=35679783
14! The damned popups lasted longer than the lesson had!
I switched over to Busuu, which has blatantly copied some of Duolingo's mechanics but at least uses them with a modicum of restraint.
It still spams you after every lesson, but I often just kill the app when it does. Quite a few ads also fail to load due to Lockdown mode or my pihole (also when away from home, due to the vpn I always keep).
I may just be their worst customer, having never given them a cent or even clicked an ad (and often not even impressions). On the other hand a bunch of people use it because of me and follow me due to having a long streak, so maybe I'm still worth keeping around.
There's a valid argument to be made that gamification helps to provide that motivation, but the argument doesn't hold up if the users aren't actually becoming proficient by using the app.
In other words, gamification isn't inherently bad, but their motivations don't appear to be good.
> if the users aren't actually becoming proficient by using the app
Learning a language to fluency requires real commitment, and I’d say an app could never possibly do it on its own. One of the most key things Duolingo gave me was consistency and a lack of an excuse to constantly practice and learn. But you also have to (and I did) use the language daily, watch content in the target language, travel and speak with locals in the language, etc. I’m not sure where Duolingo ever claimed that it alone was enough to actually reach proficiency or fluency.
Duolingo’s gamification and streaks and leaderboards gave me a reason to put a lot of effort into learning the language, and I don’t know where I’d be without it. There’s a lot of things about Duolingo I don’t love but I’m incredibly grateful that it exists.
That's changed gradually over the last few years as they switched from using gamification in pursuit of learning to using a veneer of learning as a pitch to get people to try their game.
On top of this, some people say motivation is cheap, discipline is what matters.
Instead of sticking with language learning because you have some intrinsic reason to want to learn it (or even a external one such as wanting a new job) you're substituting that with whatever Duolingo puts for their gamification. To the degree you engage with and are motivated by the gamification you are substituting your own metrics of success and progress for points and streaks.
And soon enough we end up here, where Duolingo has gamified their internal numbers and in doing so gamified your "learning".
By now my friends who use Duolingo all know it’s a game, not a real learning experience. I think they got lucky and filled a void in the market for things people think they want (learning a new language) while avoiding the parts they dislike (the effort of learning).
It got recommended by default for years when people asked for an easy way to learn a language, but they leaned hard into the path of gamification instead of trying to improve the learning experience for those who wanted to learn.
Deleted Comment
When Duolingo added that viral post on Growth hacking, it caused quite a stir about the push-notifications and gamification tactics they use. Ultimately, we decided it wasn't worth it for Coursera to veer into edu-tainment.
However, it is interesting to watch how much gamification works in adding and retaining users. In 2023, Duolingo's marketcap was 5x of Coursera. Now at similar revenue it is 20x of Coursera.
As a user, I think Duolingo is over-gamified (stopped using it) but Coursera is severely on the other spectrum where it comes off as too bland/boring to keep up the motivation. I am sure there's a happy medium to be found between reminding users to engage in something hard while doing right by learners.
Maybe this is because Coursera requires a lot more effort? My pessimistic view is that most people do not like learning, let alone learning in their leisure time with their own bucks. On the other hand, Duolingo gives more people the false impression that it's easy to learn a new language. And check out their math and chess program. They are really really easy, like pre-school level easy. Naturally, more people would be using Duolingo.
Unfortunately as they got popular automated translation services got good enough that nobody was going to pay for a slightly better and slower translation enmass.
Once that happened, that's when it seemed like they dumped their goals of teaching language and instead focused on dark pattern money extraction.
To be a bit cynical about it: the typical DuoLingo player has probably been misled to some extent about its effectiveness, yes, but also many of them don't particularly want to learn a new language. I suspect that they're happy to be able to play a popular mobile game that everyone else is also playing without the stigma of being a "Candy Crush addict" and "timewaster". "I'm learning a language!" is the welcome figleaf. https://youtu.be/F3SzNuEGmwQ?t=243
It drives me nuts that Duolingo's Japanese course does not explain grammar nor does it introduce new grammars fast enough. It's super boring to see です and ます most of the time, with occasional new grammar points thrown in. It's also strange that Duolingo introduces honorifics without context. This is super confusing. Who in the world would decipher a long string of characters out of a few really bland sentences?
I've just accepted for some time, to her chagrin, that she's effectively playing a game that just so happens to be language themed.
People always assume the alternative to Duolingo is that everyone will start a habit of reading BBC Mundo in Spanish or something, and it's obviously not true for many if not most people. And that's fine, some people are only going to scoot by with a dilettante level of interest until they take a real plunge.
Still, 3 minutes per day is just about my tempo. I don't care about literally anything in there except the learning part and consistently doing only one lesson per day makes them very nice and polite most of the time - I feel like I'm in the 'beg-these-for-money' instead of 'milk-them-dry' cohort. (Or maybe I'm in the permanent 'lets-be-nice-for-them' long running experiment?)
Similar to how you role play being an emperor in Civ: you learn a thing or two but it's no where near what the real thing is.
That's fine as a game!
Thankfully this won't happen with LLMs, as compute is too expensive so execs can't just take an easy way out of optimizing for number of questions asked
Previously I'd spend far more time in Duolingo trying to learn and hadn't really learnt any useful language skills at all. I can see it being helpful for drilling some vocab if you were learning elsewhere, but it just doesn't work to actually learn a language.
These two tactics per se are alright, right? If anything, I'd appreciate that Duolingo tries to keep me engaged. Besides, the more one spends time on learning language, the faster they learn.
The issue with Duolingo is not about gamification, but that translation is ineffective and boring, no matter how much gamification there is. Personally I find that the most effective way to learn a new language is starting with Comprehensible Input and then moving on with tons of output. Take Spanish for example, Easy Spanish, Dreaming in Spanish, Español Sí!, Extra, and Destinos offers lots of fun input for beginners. Paco Ardit's graded readers are great too.
Another problem with Duolingo is that it does not help listening comprehension at all. It turns out that we can only pick up sounds in context with tons of repetitions and combinations in consecutive sentences - a feature that is exactly what Duolingo misses. Yes, it has introduced listening and stories, but the amount of them is too little to be useful. Another lesson is that reading does not help improving listening much. When we read, we see individual words and phrases easily, while it's really hard to pick up individual words when listening. I didn't understand the difference and spent a lot more time reading than listening. As a result, my reading was at the level C1 yet I could only understand slow Spanish at the level of A2.
The timing of when I finally quit Twitter was when they shut down third-party clients, but that was after I was already half checked out because it had been in decline for a few years already (predating the change in ownership.)
https://www.youtube.com/watch?v=P6FORpg0KVo
https://www.youtube.com/watch?v=l0UE2ZY3QB0
https://www.youtube.com/watch?v=iUsDbgGQmIM
(As an aside I hate that videos are now "source material" for a discussion... it feels somewhat lame).
> wife began to complain about Duolingo feeling more and more manipulative and less and less useful.
I think this is a great bit of insight into what a lot of what the web has become! If they had been more manipulative and stepped up the quality and utility of the product would that have been acceptable to remain competitive vs something like tiktok?
A friend of mine said the exact same thing. And then a YouTube creator I follow recently made a video where he said the exact same thing too about cancelling Duolingo because he had become more addicted to maintaining the streak than learning.
Schools can give reward signals for demonstrating subject mastery, or for tuition payment and attendance. It seems like Duolingo gamified the latter instead of the former.
Things like education or healthcare shouldnt be privatized, since that always eventually ends up as profit-first game. The product suffers since milking is obvious, and quality of service is at best secondary concern.
Is is really that hard to see all this?
I found the same thing with one of the meditation apps. I was just maintaining the streak, but not getting anything from it, after about a year. I can't imagine doing that for 6 years, so hats off to your wife.
Deleted Comment
For apps, I use Clozemaster and Babbel. Unfortunately Babbel is starting to feel like it's also chasing gamification, but it does have sensible content.
For podcasts and YouTube content, I follow EasyGerman and Coffee Break German, and both are part of larger brands for other languages:
https://www.easy-languages.org/our-languages
https://coffeebreakacademy.com/
For kids… what about kids books in the target language?
Deleted Comment
Recently discovered brilliant.org, do you have an opinion about them?
This was the impression I got from the app the first time I tried it, which had to be some time before 10 years ago (not that I'd suggest nobody gets any value from it, I assume mileage varies). It just seemed to reduce an inherently arduous and deliberate effort into something that was primarily easy and gratifying. There's a very lucrative market in convincing people they've learnt rather than entertained, whether it's through easily digestible YouTube videos (a trap I've fallen into) or apps that intensely use gamification elements.
There's a difference between the concept of delivering better methods of learning or easier access to good information, and delivery mechanisms that try to make it as fun as scrolling Instagram, in my mind anyway. Coursera actually has some really solid learning paths, and free university lectures are invaluable, but even those aren't effective if you don't deliberately allocate significant time and mental energy into grinding through the stuff you don't understand, and this seems just as true for languages.
So it just always seems like a false promise to me, at least beyond the premise of making it more approachable at the outset, and I guess it doesn't surprise me that things seem to be further going in a lame direction.
I switched a while ago to Seedlang (https://www.seedlang.com/), and while it only supports French, German, and Spanish, I can at least say that the German course is everything I actually wanted from Duolingo.
Every exercise involves a real video of a real German speaker speaking in German. You get to hear them at the same time as you see their face, which is not something you'd think is a big deal, but absolutely does make a big difference.
When it's your turn to say a phrase, it records your voice and plays it back to you, rather than use some shitty model to try and guess if you spoke correctly. By listening to your own voice you can clearly hear when you're getting things right versus when you're getting things wrong. Early on, German speakers would often comment on how my accent was quite good for my level, and I think this is big part of that.
IMO Duolingo's attempt to try and scale to every language as fast as possible just makes it a worse product than something 'artisanal' like Seedlang (though of course, if there's no artisanal resources, then Duolingo might have some value to offer)
Interestingly, this is indirectly mentioned in the Linkedin post: "I've always encouraged our team to embrace new technology (that’s why we originally built for mobile instead of desktop)"
"Mobile-first" anything has always been a race to the bottom for everything: attention spans, information density and nuance, target demographic. Not just with Duolingo , but also with:
- Investing (Robinhood leaning into meme stocks and "gamification")
- Gaming (Angry Birds going from $3 lifetime purchase to a pay-to-win, micro-transaction hellscape)
- And of course, the first casualty, human communication (280 characters instead of an essay or open letter.
I don't see how one can learn German fluently using Duolingo (or even Memrise, which I think is much better). It's great for vocabulary, but I think understanding the grammar requires understanding the theory which I didn't see when I used these applications.
For me as someone who has never taken actual German courses, the biggest thing that contributed to my fluency was just listening to podcasts in German non-stop. Didn't matter if I wasn't understanding anything for months and months at the start.
I think the listening played a huge role in familiarizing my brain with wide swathes of the language. It made it so that when I learned other things later on, instead of being actually 'new', it was things I recognized and already had a sort of 'feel' for by association, even if I didn't know what it actually meant.
It was really cool watching as I went through a bit of a 'phase-change' at one point where one week I felt like I wasn't understanding more than few words per sentence and not able to actually follow conversations without looking stuff up, and then the next week it suddenly 'melted' and I was able to bridge the meaning between words and was actually understanding and following entire conversations.
My German still isn't perfect, especially my grammar and I probably should take some courses for that though. But I am at least fluent which is great.
But a book on theory can be mass produced and sold to everyone who wants to learn a language. Can't bottle and mass produce an actual experience of using the language for years. So theory it is.
For beginners to German these days, I heartily recommend the free “Nicos Weg” course from DW that goes up to B1 at least. Also has, unusually for language classes, a cast of likeable characters played by reasonably good actors carrying a consistent, building storyline throughout the lessons.
... only for me to get to Germany and realise very early on that I would need to do a basic A1 language course.
The app was gibberish; the pronunciations were wrong, the genders were misleading, and the daily interactions they tried to drill in me were far from useful.
The overinflated proficiency instilled in me by the app, made me genuinely believe I could interact easily with a German - a delusion I was quickly and painfully made aware of, much to my chagrin.
Deleted Comment
It is reasonable to expect to be a litle bit more then A2. Fluency is not.
Some techniques I use is not looking at the words when they're being read out to practice my listening (though sometimes the TTS voices make things unnecessarily difficult to understand), and I also try not to look at the word bank before trying to translate a sentence in my head first.
My main wish from Duolingo is some kind of lesson I could go into that just grabs questions from old lessons with words/phrases you haven't done in a while. It's a little too easy to get into the swing of a unit where the words are fresh in your brain's cache, but having them pulled out from cold storage would makes sure you've actually got them locked into your memory.
Also they should have a setting to disable word banks so you're forced to type everything.
Seedlang seems cool though, I'm gonna give it a download later.
> My main wish from Duolingo is some kind of lesson I could go into that just grabs questions from old lessons with words/phrases you haven't done in a while. It's a little too easy to get into the swing of a unit where the words are fresh in your brain's cache, but having them pulled out of nowhere would makes sure you've actually got them locked into your memory
Seedlang does this too. There's a gigantic library of all the exercises and you can go through them and put them in your review lists. Each time you review an exercise, you rate the exercise as 'hard' or 'easy', and depening on the rating, that exercise will then show up more or less often in the future. Eventually if the interval gets to be a year long, it'll give you the option to retire an exercise.
Each time you do a lesson, it'll list all the exercises from that lesson and you can choose which ones you want added to your review queue. It's really nice. Lots of control over your own spaced-reptition needs.
> Also they should have a setting to disable word banks so you're forced to type everything.
Yeah definitely. Seedlang also does this btw!
This exists, but not for free users.
Just ask your favorite LLM to teach you <language>.
This is not academic. All signs are saying we are heading in the wrong direction[1] and more tech ain't going to solve shit. Literacy rates and numerical abilities are going down the drain faster than you can say "Claude". I suggest we really, really get our acts together and stop trusting tech to solve our non-tech problems.[2]
“It is actually hard to imagine that every third person you meet on the street has difficulties reading even simple things.” [3]
[1] https://www.oecd.org/en/about/news/press-releases/2024/12/ad...
[2] https://archive.is/zCxBl (The Atlantic: the elite college students who can't read books)
[3] https://archive.is/4k96F#selection-1989.261-1989.387 (Financial Times: are we becoming a post-literate society?)
Heck, Latin even has a Latin-only method: LLPSI. "Roma in Italia est", you'll figure it out quickly enough.
https://duckduckgo.com/?q=dual+language+book
https://duckduckgo.com/?q=comic+book
Dead Comment
odd take, do you have the same negative sentiment for the entire field of language instruction? introductory German class at a local college? 7th grade French ? things like that?
The same thing that has worked for me as a method for learning languages has always been the same. Get books, particularly short stories or children's stories aimed at A2/B1 level, and read them. Practice grammar. Get a pen and paper and learn vocab by repeatedly writing it down. Boring but effective. And of course practice listening and talking, which means either having native friends, doing a course, using audio materials from somewhere, etc. Courses with actual humans make learning go faster (in the case of Romansh, it would have been impossible without the course).
I don't find duolingo to be effective at all, as others mention beyond the A1/A2 level. I'd be a bit more skeptical and say even A2 you need to expand your horizons.
Deleted Comment
I looked up the definition as:
> authoritarian rule, often involving the fusion of state and corporate tech power, where technology is seen as the driving force of the regime and used to consolidate control, suppress dissent, and erode public trust.
I’m trying to connect the dots between the above and Duolingo.
Duolingo has scaling and distribution. It makes no sense to scrimp for pennies on a product (e.g. English learning Spanish) that has millions of daily users. The AI radio lessons feel alienating and demoralizing compared to voice-acted stories, and the quality control is much worse.
This is classic PR spin. Do one thing, publicly say you’re not doing it. Try to get the benefit of good intentions while doing the opposite.
The CEO doesn’t lose any battles with the product managers. He could reverse the changes in a matter of days by calling a meeting or sending an email.
What’s actually happening here is they Product Managers are responding to what gets rewarded at the company, which ultimately comes from the CEO.
I run into other immigrants to this country semi-regularly now who say they've used at least one of these tools, most commonly the frequency deck or reverse-conjugator/decliner. It's been a surprisingly fruitful way to make professional connections here.
Like that would be like the Chipotle CEO proudly announcing that they’re firing their workers because they’re getting all of their ingredients from Taco Bell now due to “Taco Bell’s system being so much easier to operate” and “Taco Bell is so cheap and they have so many locations”
The ideal AI-powered tutor would work more closely to a private language tutor. It would speak with you and gradually integrating language concepts into the conversation. When you make mistakes, it could correct you on the spot and keep track of where your strengths and weaknesses are.
Actually, in some ways better. HelloTalk has a "correct the other person's sentence" feature that shows an inline diff, yet it's simple enough for people who have never heard the term diff.
When I was working as language teacher, I was tasked specifically with teaching speaking. I would often use information gap activities. These are activities where two or more parties have pieces of information but need to obtain pieces from others in order to complete the task. Sometimes, these would us language forms (re: sentence structures), but most of the time they were free flow activities. It didn't matter how "correct" the language was so long as the idea was communicated.
To think about it another, how often do we make mistakes when speaking? Writing? And yet, we still managed to communicate just fine.
That's not to say there shouldn't be any focus on form, but simply that it's not nearly as important as many tend to think when it comes to language learning.
No offense, I can tell you're neither a linguist nor a language teacher.
This is one are where the human input is invaluable and irreplaceable. Because language (the complex kind) is inherently human. It quite literally is.
However the stock did rise about 25% after his comment so maybe it was at least working for the short term if some investor wanted to cash out?
If an LLM can teach me a language, why wouldn't I go straight to the source and use GPT or Claude and customize it to my exact needs.
I feel like so many AI products these days won't be around a few years from now once more people find out that all their doing is providing a slightly different UI to what you can get directly from OAI, Anthropic, Google for cheaper and better and more tailored to you.
Because when you write it yourself, you can share it with others. Those others can then build on what you did or use it themselves.
Not everything needs to be a transaction - some people just want to make the world slightly better without asking for anything in return.
If you hate ads, deception, and dark patterns as much as I do, then most software has a negative value. But I'll only pay for it if the value exceeds the price. That means there's a rather large (and probably growing) share of the software market that I'll just never pay even a penny for, because in my opinion, it has negative value. But it's not because I don't want to spend the money. It's because I want those companies that are peddling ad-laden bloatware to be a financial failure so that the market will intervene and offer better alternatives.
Let me introduce you to the ancient lore of BrandonM, in his epic treatise "rsync v Dropbox": https://news.ycombinator.com/item?id=8863
Software freedom has never been about money, and always been about preserving public access to a software commons. You might believe your local bookstore is a better source of knowledge than your local library, but that's only because you don't really know what you're missing.
My dad likes doing oil changes himself which I'd never do, but it doesn't occur to me to insult and question why a person has different values than I do.
Deleted Comment
Duolingo being crap and costing anything/month is the problem.
(Paid Duolingo user and long time GNU/Linux user here)
Anyway, Im responding to you because i find ChatGPT to be a FANTASTIC tutor. Like I was absolutely blown away. It can do all the translation stuff, but also answer questions about different verb tenses, conjugations, syntax etc... I'm sure that an extremely good spanish teacher would be better, but I think ChatGPT is probably better than most. And it is free.
When I am learning something new in development, the LLM is immensely useful since it has unlimited patience, and I can zero it in on exactly the level of complexity or understanding that I need.
Dropped it instantly. I get the bargain-basement cost-cutting appeal from a (bad) CEO's perspective, but if I'm paying actual money for a service, I want said money going to humans.
And as others in these comments note, if one wants to use an LLM as a language tutor, you can do that for less than what Duolingo is charging while also getting the benefit of being able to tune it to your exact needs instead of being stuck with whatever Duolingo decided is best.
This is what I don't think the "AI-first" business crowd understands -- in many cases, the moment you admit the humans in your organization can be wholesale replaced by AI, that's a sign it's possible your whole ass business case could be unnecessary LLM middleware.