It's quite hard to lose a Duolingo streak
leejo.github.io
leejo.github.io
Then one night I started reading at 11:57 PM, and by the time I was reading for 5 minutes, it was 12:02AM and I technically missed a day. I didn’t even realize how important the streak must have been to me, because I stopped reading after that and haven’t picked up the iPad to read a book in almost 2 years since then.
Edit: to one of the flagged replies that assumed I wasn’t reading much anyway (which I can’t reply to)… to be clear I didn’t just read five minutes a night, that was the minimum amount to keep the streak going. I read (iirc) 8 books in that time frame, including the lord of the rings trilogy in there. I wasn’t reading for hours a day, but probably averaging more like 20 minutes. Not a ton, but it’s not like I was barely reading.
Non-avid reader before to almost a year of avid reading, then all lost because of a hook metric. This stuff is a real deal.
Then, like GP, I just had a particularly busy day and didn’t get to my reviews until after midnight. BAM- a triple digit streak gone. After that, every time I logged in I got a reminder of my FAILURE (real or imagined) to keep up with my studies. I didn’t like being reminded of that, so I didn’t login. I ended up taking a year-long sabbatical from language learning instead.
Developers, don’t put streak counters in these sorts of apps.
If I were in your shoes, I would do Memrise to fill the little moments throughout the day. But the focus of my studies would be (1) a local community college Portuguese course; (2) the FSI Basic Portuguese course (or another drill heavy course from the 60’s, maybe linguaphone?) for self-study drill practice, 30min a day; and (3) a good sentence deck in Anki, minimum 10 new cards per day.
SRS apps can be good for building up a vocabulary base. Self-study drill based programs like the old FSI series can be great for internalizing grammar rules so that you don’t have to consciously think about it.
That does give you a great foundation for developing linguistic fluency though, should you relocate to the country or start immersing yourself in native media.
There's a streak, but it's so easy to ignore, specially since the app focus on 'learned words', something that duolingo had many years ago. The study is also organized in just chunks of 100. So instead, I focus on how many words I want to learn on the next week/month, or when I'm getting to the next group.
This definitely deserves a deep discussion. Get to the bottom: is it good, bad, is there a better way to implement?
I can speculate that, for many people, just getting a triple-digit streak is a big win.
Even if they drop for a year, or even forever, after losing it.
They wouldn't have accomplished the feat without the streak metric.
I don't have any data. As I said, it's speculative, but perhaps reasonably possible to be the case?
The trick is 1) finding the story that you want to read.
and 2) putting in the time to reach that state of flow/pull.
It's extremely pleasurable and I wish it on everyone.
I think the timetracker sort of motivation is like the runway before the flight. The bootstrapping. Or the kindling of a fire that will grow to sustain itself.
Also, days are like 48 hours long from the first timezone to the last timezone.
There's also the cracked lesson feature. They aligned cracked lessons to urgency gamification instead of something more sound like delayed repetition. As a result, you'll often have cracked lessons which you've long since permanently committed to memory and will be spending 15 minutes on easy lessons rather than doing the important, harder lessons.
It seems that the better language learning app would rather ask you to use the app less, if you make more progress when you do use the app.
The collection of recently failed flashcards that you can iterate is a great feature. It is somehow an automatic feature of spaced-repetition apps, since recently flashcards are shuffled into the stack at a higher frequency.
You _will_ need to supplement the vocabulary with Anki or another tool. You won't become fluent, but you will get a good basis to start talking with people comfortably. Becoming fluent depends on keeping up that talking. I would say that Duolingo gives a terrific starting point, at least for the three languages I've learned with it.
From my point of view, you can make it to A1/A2 proficiency in a language with Duolingo alone - after that, it gets quite hard. Teaching grammar and conjugations isn't its forté, without resources outside of Duolingo it will get rather tricky.
That being said, I still use Duolingo for my daily dose of language exposure - but without Anki (for vocabulary), youtube videos (to explain various specific topics) and evening courses (which help a lot by giving a structured approach on what to learn in what order), my learning curve would probably flatten out.
After nearly a year, I can help set up for dinner ("Where are the plates?" "There is bread on the table already" "Do you want tea?") and similar things.
Now I'm getting into genitives and such, and that's where Duolingo really seems to not do well. It doesn't explain why a word appears a certain way in a tense, it just makes you memorize that it does. Sometimes I'll ask my husband why, and he just says "Well, because that's the way it is". Of course I'll use the same explanation when my MIL asks me something about English.
Most native speakers are useless for why. You ideally want someone who learned both your native language and the language in question. Then they can often explain things relative to your language, and you'll learn things about your language and the language in question. Look for language notes from Hong Kong; I had found a university site with very helpful information on French many moons ago, but I can't find it again, it was written in fluent English, but with insights on English usage that were unlikely to occur to native English-only speakers.
I did French all through high school and most of college, then learned firsthand that I was maybe 25% as fluent as I assumed. Then I went home and went through the entirety of the Duolingo curriculum in a few months, without learning anything new (a couple new vocab words, maybe).
So, you could view this in two ways: either it's a scam that tricks you into wasting a lot of time viewing ads or paying a subscription without delivering on its promise. Or, you could see it as a shortcut that's worth at least a couple years of classes. But, at least in my experience, it does not enable you to speak a language with fluency.
(Which makes total sense if you step back and look past the marketing promises. Of course 20 minutes a day with a cartoon bird is not going to replace actually living in a country and being immersed in the language 24/7)
key problems:
1. not enough speaking practice. speaking is _at least_ 50% of the skill but about 10% of DL.
2. Waay too much time spent setting up each question w animations etc. not enoimugh learning time. much more repetition is required.
3. way too general learning. people have varying needs for language. e.g. work, travel, family, research/reading. teaching must be tailored to these.
Else you're like a dog listening to steam of human sounds: "Not walk, not walk, not walk, ...WALK!! (I know that one!), not walk, not walk, ..."
In my experience, Duolingo and other entry level apps can get you there especially for the person who hasn't learned a second language thus doesn't even know where to begin.
Not in my experience.
Anki + language-partner sessions + evening classes did it for me.
Luckily, that is a lot easier these days with the internet!
I can still understand a lot of Spanish and when I’m warmed up and have been back a few days my conversational level comes back.
Spaced repletion algorithms can be really helpful, but they can also be a hindrance. Duolingo treats the user like they don't know better than the algorithm. The language learner needs to be the one in the driver's seat with the algorithms and course material being there to support the learner.
Gamification is great, but it should only be there to help the user stay motivated meeting their goals. Like you mentioned, if you've already mastered something having it presented in your study session is actively working against meeting your goals because you're spending time studying something you don't need to.
I don't see studying everyday as an end-goal, though studying every day can be helpful for retention. You should feel rewarded but not punished when you "loose your streak".
Focus on the days, and the streak will take care of itself.
Focus on the streak, and one day you might realize that it hasn't been worth keeping for a long, long time.
Put it another way, is 10 years with a few lapses worth significantly less than 10 years completely sober? Of course not. It's still a huge achievement. So why punish yourself?
If you think it's worth lying about having 10 years, then why bother lying about having 10 years?
In practice, milestones are more for sharing one's experience than for attaining an award. And if you slip up, telling that truth is honestly more helpful for the other person struggling to stay sober than lying to them about having 10 years.
You need both.
The ability to not lose it all because of a break.
But also a reason not to have a break.
Cf: World of Warcraft
Many important things in life like fitness, diet etc are very repetitive and hence boring to some extent. There are times when you just to have to get through the grind even if you are not interested or you are aligned with it. The point being don't just depend on the initial euphoria of things to do a thing that's good. If something is good on the longer, at some point it does get a little boring, and that's ok, just keep doing it.
There are things that I stopped doing because I was bored and a few months later I had like a moment of epiphany- It wouldn't have hurt me if I had stuck to this, quitting has cost me more.
May be one should think beyond doing things for immediate emotional relief.
There's also the fact that your lower brain systems don't care that you "know that is is meaningless" -- they still see that number getting bigger and give you that dopamine hit, which is what will really keep you coming back for more.
Like, when I have to travel to visit my family, it's like 25h of flights and connections + another 2h road trip, and still need to arrive like 4 hours earlier. I'm def. going to be offline and would miss a streak if it required me to do every day. So I would not even try to increase it anyway, since I know it's impossible not to lose it a couple times every year.
Since I know I can go offline for a day or 2 when necessary, I have more motivation to just keep going
But as soon as my streak ended, my frequency of playing went from 6 days in a week, to 5 days.. and now I haven't played it in months. Fake or not, keeping the streak alive did help me learn a lot of French (well actually there is one more side to it.. even after all that I can't speak a single proper sentence now apart from the basic je m'appelle stuff)
This is the real problem with Duolingo: it doesn't actually teach you to communicate in the language. Rote memorization and communication are two different things, and Duolingo does very little aside from drilling memorization. Streaks and gamification don't contribute to communication, which has its own system of rewards: when you communicate successfully, you accomplish some real-world task like buying yourself bread or getting a date, and that accomplishment doesn't need game-level rewards. Gamification is only effective when trying to palliate some sort of otherwise boring "grinding" (to use a gaming term from RPGs).
I think it's too dramatic to say it's a "real problem" with it! Anybody who intends to become fluent in a language is going to need to practice actually speaking and writing.
Is rote memorization essential to language learning? Absolutely. Should it be the focus of your learning after the very early stages? Probably not. It's just something you do, all the time. Memorize all you like, but true acquisition only comes after you've used your words in real conversation / listening, so that's should probably be where you place your effort. But that doesn't help Duolingo.
In the Japanese learning community, the seduction of this kind of rote, mechanistic, "streak-y" learning has been a trap for a long time -- so much so that there's a popular methodology that has people spend a great deal of time "memorizing kanji". It's appealing because Japanese has a big alphabet of complex characters and the methodology suggests that you can turn learning into a simple memorization exercise with discreet, measurable progress. The problem is, it's pretty much orthogonal to the act of speaking (or reading, or listening to) the language. Learn 2000 kanji and you might gain some ability to infer the meanings of words, but you're still essentially illiterate. The same thing applies at the level of words and sentences -- I have lots of words that I "know" on flashcards, but aren't available to me in actual conversation.
People get absolutely addicted to this stuff, spend years on it while gaining no meaningful level of language skill, then get frustrated and quit. The tragedy is that most people could put in a fraction of that time engaged in listening and speaking with a simple vocabulary, and end up in the kind of virtuous cycle that actually leads somewhere.
I’ve never tried it but I feel like it’s probably similar to Spanish classes in high school.
Duolingo doesn't put so much emphasis on conversation in my experience. Most of the questions give you the words in one language or the other and have you write out an answer. This allows you to lean on reading. Listening and speaking are the hardest parts so the should have the most practice, not the least.
If it's too easy, you can make it a bit more challenging yourself.
(I suddenly found myself stuck in Brazil for work for two months this year, and after a couple of hours of DuoLingo I was able to exchange a few phrases with hotel staff, often interspersed with the most useful phrase 'O que isso significa?' (What does that mean?')
The best part of DuoLingo was (and is) to save my conversation partners much of the 'Oh, in Portuguese we call that...' we'd otherwise be doing.
This is a more general problem with gamification and external rewards: they are just as good demotivators as they are motivators, if not more so.
This is why I wanted to turn off streaks when I was using Duolingo. I knew it was just a matter of time before I had my streak broken, and I was very concerned about what would happen then.
The gamification requires two coordinating motivations— secondarily the game mechanics, and primarily learning a language. Duolingo won't replace candy crush for non-language-learners. I imagine folks significantly discouraged by poor game performance without doing some serious learning first just aren’t motivated enough to practice their language skills in an app.
Personally, I’m not motivated by gamification at all and am content to ignore it. It is noisy, however, and I also wouldn’t mind the ability to shut it off.
That said, if you want to learn a language you need to study a lot. Complete everything Duolingo can in a few months and then find better courses of study to learn after that. You should never in my opinion have more than a 6 month streak.
It's similar to redit where someone demanded a subredit of mine since they deemed I not active enough (I was active but not to random guy's satisfaction) and the redit mods agreed. I haven't been on redit since after being a member for 15 years.
So yeah streaks, the chain (as Jerry Seinfeld says), consistency being broken makes me feel like crap.
But they decided one day to just nuke it... for no reason as far as I could tell. My subscription went with it. Shame.
EDIT: based on the responses below I did a quick search and it appears that it was silently removed, refactored and then returned slowly back to people depending on their account age/device. This is actually incredible news (to me).
First day of use: 11th Dec 2014
First missed day: 19th Jun 2017 (921 days later)
First freeze use: 20th Oct 2020 (2,140 days after first day of use)
I think when I started using the freezes was when the value of the app had dropped for me. 2,140 days is a pretty good run.
FWIW: living in a French speaking country has provided the most value. Duo helped initially with vocab building. I find/found the lack of colloquialisms one of the biggest problems with it.
This is actually a big problem with how people perceive language in general. It's hard to understand if you only speak one language, but the idea of a language as a discrete unit is a myth. Here in Québec there is Québécois French, which in and of itself is a generalization, as many regions have different accents and expressions.
In France there are many kinds of French. Some people say they "can't even understand" people from a certain other region... That's pretty much where a dialect crosses the line into a language!
In general people will learn what we call "international French", which is spoken by no one but understood by all. But good luck navigating the subtleties of a French-speaking society without being familiar to their regional dialect.
Sure, someone with a strong uncommon accent can take some effort to understand ("can't even understand" is a clear exaggeration except in the more egregious cases), but the language underneath is the same.
French is very standardised, and "international French" is spoken by everyone. That description would fit something like German much better.
You sound clueless. The majority of people don't speak "international French". Do you even speak french ?
Do you have a sense of what kept you going? And what about Duolingo helped?
That flexibility has helped me track my streak outside of the gamified experience that is Duolingo. It has helped me stay motivated based on my personal goals and not be artificially motivated by the whims of the App developer & PM who are focused on driving goals such as DAU/engagement and monetization.
The ability to make up the missed days might have led me to carry on. Or maybe forgiveness if you'd gone over in the precious few days.
It's sad really it was quite an accomplishment.
Also, not getting in enough stand hours because I have to put the watch on the charger during the day (the non-replaceable battery of the watch is on its way out) was another reason to stop chasing rings.
I still use the watch to track my activity and my sleep but I gave up chasing badges.
... Pavlovian conditioning works, even when I'm the rat. For the time being, I'm okay with that, because Duolingo does help me learn French, let's see how it goes long term.
Can you say more about how much it has helped you, in what ways, and what else you have been doing to study?
Once you go into the B1+ area, the usefulness of Duolingo starts to fall off, slow but steady - learning grammar and conjugations in all the different tenses isn't its forté. I think it's still very useful for repetition and constant exposure to the language, but without content outside of Duolingo, I'm not quite sure if it would work out so well.
For me personally, YouTube has been an amazing learning source for French - there are a lot of channels out there with tons of high quality teaching videos. If I had to name one, "Learn French with Alexa" is probably my favorite, but for a lot of topics it's very useful to watch multiple videos from different channels on the same topic, both as repetition and also because sometimes having multiple, slightly different explanations helps.
Since a few months, I do visit a weekly French course though - self study brought me only so far; reading already works quite well, listening so-so (depends on the talking speed), but I have a hard time speaking myself. It takes a very long to find the words, but the more I have to talk in course the better it goes.
One more thing: For purely learning vocabulary, I very much recommend Anki, it's amazing to just memorize words.
Ugh, this got a bit longer than intended, hope it gave you a bit of an overview of my approach :)
There is a lot of diversity in instructor-led training. None of our classes are anything like what you would have experienced in highschool or college. Our largest classes have 4 students in them, and that's strictly a cost-cutting scheme that Gov has insisted on for maintenance training. You'll be one-on-one with your instructor. They'll be a native speaker with often decades of experience teaching language. They'll be tayloring a lot of supplemental material specific to you. We even have some of our own apps that we've developed (that's my job), nothing like Duolingo (though nothing you can just use without our curriculum or instructors).
A lot of people say "the best way to learn a language is to live in a country where it is spoken". Clearly, that is unrealistic for the vast majority of people. I would also say it presumes you are a specific kind of extremely extroverted, fearless person. Most people find themselves becoming quite introverted in situations where they can't communicate easily.
For these reasons, I think the most realistic answer for the best way to learn language is private tutoring. The OP take about having used Duolingo for almost 10 years. That seems absurd to me. I would hope they have mastered at least half a dozen languages with that kind of effort. Language training shouldn't take that long.
For people who have a specific need to learn a language this is of course true. Or even for people who have an amount of time and effort they are willing to dedicate to that end.
The thing is, though - I’m very unlikely to invest in private tutoring in order to learn a language speculatively. The time and financial requirements are too high a barrier.
But Duolingo is an investment of maybe 10 minutes a day at some moment of downtime. I’d estimate that I’ve spent maybe 200 hours over three years with it, which has taken me from not knowing a language to being conversationally fluent in that language - something I would otherwise just never have done.
Duolingo (or other similar systems) are very obviously not the optimum way to learn a language to fluency. And it’s especially obvious that they wont replace a tutoring service for government and military clients. But a lot of the criticism of them completely ignores the fact that they fill an entirely separate niche which is totally valid.
This really jumped out at me. 200 hours to get to conversationally fluent is very, very fast: in the CEFR framework, with proper tuition, this generally gets one to about A2 ("elementary").
How closely related is the language you're studying to the language(s) you already know? Do you supplement Duolingo with other resources? And, last but not least, what's your definition of "conversationally fluent"?
But you can't have a private tutored lesson while in a queue at the supermarket. Or on a train. Or while you wait for the water to boil.
Of course the government pay for your services. They need people to be trained to be good at languages. Most of the people who use these apps are doing it as a hobby.
Eh, the implication in my post is that Duolingo is minimal effort and the streak doesn't really mean anything other than you may have maintained that minimal effort for a long time.
Besides, very few people are mastering a single language let alone many. I'm a native English speaker and sometimes I feel I haven't mastered that language.
And yes - private lessons are going to offer the most value if you can't live in the country. Even then it's only in this particular part of this country where I would hear something like "On prend le traclet pour acheter du Papet."
It's not nearly enough on it's own to learn a language though, it's very very good as a supplement though, ime
Based on my experience of language learning, I agree with you.
> The OP take about having used Duolingo for almost 10 years. That seems absurd to me. I would hope they have mastered at least half a dozen languages with that kind of effort.
This completely glosses over the costs involved.
Has anyone had success with finding tutors on platforms like italki.com? I tried for a bit but probably didn't spend enough time to find the right tutor.
italki - yes, absolutely, but taking your time to do the research definitely pays off.
My method was to shortlist teachers who:
(1) fit my linguistic needs (country/dialect, native speaker status etc); and
(2) satisfied certain availability criteria (timezone, teaching days/hours etc); and
(3) had a high enough number of students and -- critically -- a high enough total_lessons/total_student metric (I forget the cutoffs I used).
I then looked through their profiles (there were by this point only tens of teachers left on my list for German), arranged trial lessons with a handful, and picked the one.
Been with that teacher for a few months. Couldn't be happier with the experience and with my progress.
Forgive the digression, but this reminds me of a general gripe: I wish there were more/better introductions and tutorials and other learning materials geared toward people knowledgeable in another human or computer language, the educational equivalent of P frames in temporal video compression. Rather than feeding you examples piece by piece and making you build up the mental model tediously, brain dump the mental model on you and let you fill in the specifics as you go. Easy Arabic Grammar by Jane Wightwick or Basic Polish by Dana Bielec are decent examples of what I wish were more prevalent.
As someone who has used Duolingo for multiple languages (German, French, Swedish) for about a decade (2012-2020) I am here to say that Babbel is vastly superior to Duolingo in terms of quality.
Duolingo is just rote memorization of their vocabulary. Despite investing years into "studying" the above languages in Duolingo, I didn't feel I was actually useful in these languages (I studied before relocating to the respective countries).
Babbel on the other hand, is very much like a digital version of the language textbooks you will find everywhere for levels A1/A2/B1/B2. The Babbel lessons involve listening comprehension, speaking (though their voice recognition is... poor), reading, writing, and they make a huge effort to teach you conjugation and grammar rules. There are also no ads or annoying animations.
Babbel does cost money, but it's far less than Duolingo's paid tier, and I personally find Babbel much more effective at actually teaching me the language as opposed to Duolingo having me memorize random vocabulary.
Babbel allows you to set learning goals (though max target is 9 per week) and daily reminders, but they don't have streaks nor do they get up in your face about it.
I will never go back to Duolingo after having used Babbel.
There are also many similar gripes among STEM tutorials. Why can't this physics/chemistry/economics book assume I know vector calculus/differential equations?
You should never be in a scenario where you're using Duolingo without learning anything new unless it's intentional or you've mastered the entire selection of lessons they offer.
I did use Duolingo a bit when starting out, but I felt like I didn't really remember any of the words introduced there, just learnt how to hit the buttons as fast as I could. (And oh gods those annoying animations). Pimsleur on the other hand is amazing when starting out. Then a combination of podcasts, simple stories and Anki.
I've also heard good things about italki, seems like a marketplace for 1-on-1 language tutoring online.
He also surrounds the verbs in mock clozes, so you get hammered with context. You can also do it without doing anything else; the Romance verb situation can be isolated from the rest of the language completely. At least for Spanish you'll eventually need to know what common subjunctive triggers are, that they'd rather use the present tense (with context) than the future or past more often than you'd think, and other stuff that's an order of magnitude simpler to remember than those damned verbs. If you need to know irregular vosotros conjugations in subjunctive future, you've already got them in your head.
http://www.asiteaboutnothing.net/w_ultimate_spanish_conjugat...
But of course, when you're doing that, you're pressuring yourself to get through as many lessons as possible, not to engage well with the material. You don't want to stop and read the discussions about a particular translation. And (since I'm studying Chinese), you often have a choice between attempting to speak Chinese, typing Chinese using pinyin, or clicking on the character blocks. The first two are almost certainly better practice; but the last is a lot faster and more predictable.
And of course, now when I'm out of that rhythm, I'm totally not motivated to study, since I'm getting half the points I would otherwise be getting. If I'd been rewarded (not competitively, but just personally) simply for spending time in the platform, I'd probably be spending it more effectively.
I'm rather pissed if the daily reminder comes to late. If I don't want to practice daily anymore, I'll simply switch off notifications or uninstall the app.
https://dev.languagereactor.com/phrasepump (this is the dev site, will go to prod in a couple of weeks)
The interface needs work to be usable on mobile, but I guess you know that.
I'm not sure whether the TTS you're using is really good enough to be usable for learning a language. I tried it in German and noticed quite a few TTS bugs, like pronouncing "Tu" as if it were an acronym "TU" instead of the verb "Do", as well as switching to an English pronunciation next to English names (pronouncing "Hallo Roger" more like "Hello Roger").
Tatoeba also has many recordings in quite a few languages (e.g. German https://tatoeba.org/en/audio/index/deu ) You should try using them if possible. For download instructions, see https://tatoeba.org/en/downloads
Edit: typo
I started working on some of my own tooling to help with this. For example, one tool took a list of sentences and translations and ran them through Google TTS to get recording to use in Anki. After working on this tooling for a while, I decided to move away from Anki and just build an app.
Right now, we're working on building out a beginner Thai course focused on reading because educational material in that area is severely lacking. We're also working on a series of graded readers.
Soon we'll be launching quick-study feature that allows you to curate lists of words and phrases that you want to learn. You will be able to select from anything you've already learned, draw from our existing library of 129,000 words and phrases or input new content. Once you've created or lists you can use the quick-study that will create a custom lesson for you based on how familiar with the content you are.
Not launched or even in beta yet, but the site is public. The info is listed in my profile and you can contact me from the site if you'd like to get in touch.
One of the happiest days in my life was the day after losing the streak on 750words a few years ago.
I had a 500+ day streak going but it had not built into an effortless habit (like a cold shower).
Surest way to misery is feeling forced into doing something yet not getting enjoyment nor showing any improvement.
I am reasonably sure if I had tried to not break the streak in say doing standup ala Seinfeld the results would have been similarly dismal.
On the same note I quit Duolingo when I realized that the modicum of German I was learning was quickly overtaken by my teenage daughter. She was in a very average German class spending less time on lessons.
how much german have you learned since quitting?
I find it overbearing that apps create the expectation of participating even when I am in the middle of the forest and mountains with zero network connectivity. It incentivizes unhealthy lives and life patterns of never disconnecting from the internet. Why can't these apps allow users to choose what kinds of streaks they want to aspire to?
Do it because you want to, you are getting something from it, not because you have to.
Ended up with a ~400 day move streak. Yay me, right?
Well. It got pretty annoying toward the end, having to wear the damn thing every day. It kinda creeped up on me, one of those things where the motive to just not wear a watch for a day is low, so why not put it on and keep up my streak?
But cumulatively? There's a huge difference between mostly wearing a watch and always wearing a watch.
It was a relief when I busted it, finally, for two reasons: I could have a normal relationship with the device on my wrist, and I was in no danger of ever setting another record streak.
After that I deliberately bust streaks in apps at a number I choose in advance, something like 50 or 100.
Another issue that caused me to lose my streak is when Duolingo says: "Sorry we encountered an error. Don't worry your streak will be safe." In my experience my streak was not safe.
I don't mean to be too harsh with Duolingo, I think they get a lot of features just right. The streak thing, however, was just an annoyance for me. These failures made me temporarily quit several times. Even if you know you should not become upset about stupid internet points it still hurts when you feel it is not your fault.
Far more people like the illusion of learning than actual learning, so it makes sense from a business POV, but if you're really dedicated, textbooks + flashcards + immersion will get you farther ahead.
Apps like Duolingo and others also completely neglect giving the language learner longer passages of text to read or videos to watch. Most learners do this on their own, but it does take a tremendous amount of prep work to find the appropriate materials.
Once you have the materials, you need a way to re-enforce what you've learned. I was doing this with Anki and found it very cumbersome and time-consuming to do.
Anyway, I try not to look at the points or streaks, my real motivation is to learn the language.
Drop Duolingo, use a proven method like Pimsleur, grab the Anki Core Vocab deck for your language, and start consuming native content ASAP.
What's more, it's easy to guess many answers knowing nothing. Simple things like the only word that's capitalized will be the first word in the answer.
I also remember thinking the vocab for a given topic was really arbitrary too.
My problem with Duolingo ambiguity is that if you want to do language learning on Duolingo starting from the middle rather than at the beginning, you'll get a lot of questions wrong because you're not familiar with the exact vocabulary that Duolingo wants you to know (i.e. what it taught you in earlier lessons.) Since it's constant translation drills, it builds up its own internal conventions about what English word translates to what target language word that you don't know as someone who has learned from other sources.
I've found translation to be a dead end for me. If I don't know a Spanish word I look it up in a Spanish language dictionary and/or thesaurus, and if I can't manage to figure it out using that and context, then I'll finally look for an English translation of the word. Totally there for English language grammar instruction, but not translation.
I don’t want animated cartoons emoting — often incongruously with the sentences, but even when the emotion suits they are grossly exaggerated — in response to correct answers. I don’t want the “motivational” messages that remain even after disabling motivational messages. I definitely don’t want the “motivational messages” toggle to randomly reactivate, spamming me with far more quantity of annoying messages interrupting my flow. I don’t want the news feed. I don’t want the “lesson complete!” screen, followed by how many skills I’ve restored today, followed by my position on the league table if that changed, and possibly one un-skippable animation of a chest opening for each of the Daily Quest goals I met.
I do want the quizzes to load quickly, and (when on the web) to be openable in background tabs so I can queue up all the specifics I want to revisit in a session — years ago now, I did that with the entire German course, revisiting each lesson once over two days.
But you’re right, they don’t care so long as people keep coming back.
"your best time is now 41sec (instead of 42 2 years earlier)" -> that's when I stopped completely using it.
"Ugh... they implemented this for FOMO... but damn I can't skip playing this game for a day because I am going to miss out on the +1 bonus to agility that's gonna be awarded if I log in 7 days in a row."
Duolingo is pretty good for getting you started with a new language, and can provide some valuable practice, especially if you don’t yet have a teacher. But don’t take it too seriously. Its translations are often poor:
No. Probably five years for one and then switching between others (German, Italian, and Spanish).
> What % completion are you for a the language?
I got close to the end initially then they added more to the tree so I never finished the tree as I didn't need to. I just looked at it and I'm at 50% so they've added a lot since I last checked.
> Do you consider yourself proficient?
I've lived in the French speaking part of Switzerland for nine years. That has made me proficient. Having French speaking colleagues, going to evening classes, giving talks in French, receiving correspondance and filling in my tax returns in French, working in a local gallery (adds vocab and situations outside the norm), watching French films and TV, listening to French podcasts and radio.
Duolingo helped with building some vocab and grammar.
Becoming more proficient in French and living in a French speaking country has, on the flip side, started to erode my English (spelling for the most part).
*once you have a one year streak, you become VIP which means you can 5 days of "vacation" instead of the usual 1. so thats 5 days that recharge once a month. In order to present the number of streaks that this guy has in his screenshots you will need to spend "gems" or "money" whichevever comes more easy to you.
So yeah, its nice that after investing time every day for one year, i get to get a week off in between ;)
However, I am not sure if there's a use case for it that Anki doesn't achieve with notably less adverts. Am I just not the target audience?
It usually happens when you do two or three lesson in a string, never seen it on the first exercise of the day.
What I miss is a sort of weekly recap across all units.
Barring Gboard, all Android keyboards that I've tested do not support automatic switching of languages[1].
Even with Gboard, you need to explain to the user how to install the language for the keyboard. This is not trivial.
You then may have to deal with another input method, and recommending one to use (for example: do you want Pinyin/Cangjie/Zhuyin). This is different for each language.
You'll probably need to deal with Android manufacturers doing ridiculous things to lock down their devices. I would not be surprised if some restrict you from changing the default keyboard.
And if you make your own virtual keyboard, you've got a lot of accessibility to think through.
[1] https://developer.android.com/reference/android/widget/TextV...
That is the kind of thing I’m talking about. Just not challenging at all.
The second lesson was a story which was a lot better than what I was seeing before, but I still think they adjust the difficulty so you are missing fewer questions so that you’ll stick around.
Can you hold a conversation in said language?
I understand because of that I am not their target demographic. Or maybe I am and they want to engage users like me more. Maybe if the UI felt a bit more immersive--a clip of native French speakers interacting rather than cartoon bubbles and gems, for instance.
I do realize I can Google it. But isn’t it interesting that Hacker News brings all of us into contact with alien civilizations every day. Whatever Duolingo is, sounds like it became a significant part of you life. For me it has been Wordchums and FastHabit.
My comment was merely a bit of musing about how big the world of ideas is. Don’t you also find that reading Hacker News you stumble on entire ecosystems you never heard of before?
Anyway, I did look it up. Apparently it’s about learning languages, in case you also didn’t know.
B) I travel a lot and I realized that I don't actually want to "learn a language" I want to "be able to communicate in an area". So doing a Duolingo speedrun on the airplane is pretty much useless (or even months before). What you really want to do is know how to buy a girl a drink, for example. In a loud situation this may even involve counting to the bartender, "language learning" apps aren't going to show you how to count on your fingers the local way. It really changes the priorities to what you actually need to use.
The more stimuli the better, because the less actual talking is expected. Very basic bits of information need to be transmitted.
Nobody is going to understand your years of app language, or years of school language, and definitely not your accent. You need idioms and expressions and you aren't going to blend in either way. Do it live.
I think there are many people that also really just want to "learn to communicate in an area" but think that requires "learning a language". for that group of people that also has the time and privilege to go to the place, use my strategy
for some other group of people, have fun telling your phone how a "young girl eats an apple" for 20 levels.