some people are quite fine learning a limited number of phrases to lurk in a country. a great part of communication among humans also happens with the body/eyes. no one needs to discuss their phD dissertation in 4 different languages
[0] https://blog.duolingo.com/can-duolingo-make-me-fluent/
edit: Duolinguo also is nice (and make a funny non-invasive joke) if you are using something like uBlock!
This happened to me about ten years ago.
I too had not bothered to understand pedagogy. It is only when I wanted to learn Sanskrit, and struggled with it, that I got pissed off at the lack of progress and began looking around. There are some people on YT who talk about this stuff:
- Alexander Argüelles
- Steve Kaufmann
- Luke Ranieri
I might be missing a few others.
You first have to know what your problem is, before you can solve it.
> no one needs to discuss their phD dissertation in 4 different languages
True. In culturally homogeneous countries, you don't need four languages to make yourself understood.
It becomes somewhat necessary in places like mine where different groups of acquaintances/relatives/friends speak different languages and finding a single language at the intersection of those groups can be hard.
is that Duolingo fault or users? because that happen in any hobby. heck, take indie gamedev.! hundreds give ups for a single released game. we could also say that there are people who tried Duolingo and years later they are fluent because the app was the kickstart
you have to be quite naive/lazy to stick ONLY with Duolingo for a year or 2 and expect that you will be fluent. there's also different ways of approaching the app... like each lesson allowing one to read or discuss it with the community; meta-thinking stuff like "am i learning or just rushing through lessons?" etc.
i heard podcasts about psychologists suggesting that fluency is subjective and it happens at +4 years time span of active engagement after mastering the basics
i think Niklas Luhmann’s essay on communication is quite relevant here; https://www.unisalento.it/documents/20152/2157613/LUHMANN-Wh...
I have seen lifelong scholars of the Sanskrit language struggling to speak in Sanskrit because they are simply not used to it.
That's true, but the opposite extreme can be even worse. In YouTube and Reddit I see so many people procrastinating in their quest for the perfect learning method instead of just sticking with any of the good enough methods they already have. I know because I've also kind of fell for that trap myself sometimes.
In fact, I imagine that the average Hacker News user is far more likely to fail at language learning because they procrastinate on linguistics and pedagogical theory and not because they churned 10,000 hours at a slightly suboptimal learning methodology.
For some people, it is because they are unsure of which method works for them. So they wander from one theory to the other.
The rest simply enjoy the meta aspect of the journey more than the journey itself.
Sure, they've got that fig leaf covering them.
I don't think that Duolino is absolutely useless. But my reason for learning languages might be different from those of others. Some people want to be able to say a few words in the language they are learning. I want to read novels, poetry and philosophical texts.
The approach you take and the kind of vocabulary you want to acquire will differ accordingly.
It works by estimating the difficulty of English sentences, then translating ones at your level into your target language.
I'm curious what determines whether or not you add a given language to the list. DeepL and Claude, at least, have usable translation ability in more languages than the app currently supports. Is there a lot of manual effort required for each language, or do you want to keep the list limited just to avoid overwhelming users?
DeepL is actually pretty limited in what it supports. Unless I've missed a new language, Nuenki supports all of DeepL's languages.
Some of the additional ones are supported via Claude only and, where permitting, Groq. Groq is far faster than Claude; in languages that DeepL supports, DeepL handles visible text and Claude handles text that you haven't scrolled to yet. Claude-only languages are a bit of a worse experience.
It's pretty easy for me to add a language. It's all stored in a centralised toml file, which happens to be open source - https://github.com/Alex-Programs/nuenki-languages/blob/maste... - and it's about a 20 minute job to add a language, test it, etc. Then it's about half an hour and 5 USD to benchmark whether Llama is any good at it, and if so enable Groq and make the experience a bit more pleasant. I'm currently working on improving the translation quality benchmark (https://nuenki.app/blog/the_best_translator_is_a_hybrid_tran...), because people seem to like it and there's definitely a lot of room for improvement.
That 20 minute number is without updating the big language cloud on the website, which is a bit finicky; iirc I haven't added Vietnamese to it yet.
If anyone here has any requests, I'd gladly add them!
My biggest progress in English was when I started to read the English internet (HN, Reddit, etc.). I used an browser extension to translate words that I didn't know.
I'm learning Spanish now, but there is no content that interests me. Maybe the Spanish Wikipedia sometimes.
So this extension gives me that language exposure.
Suggestions: Can I get some transliteration when I hover over it, rather than translation? Maybe a Alt+ ?
For some languages it's available but disabled by default. You can change it in the extension popup under "Hover Behaviour".
In Danish the third line is translated as "Du kan vende tilbage til oversættelser ved at holde musen over dem." which means "You can return to translations by hovering over them." i.e. the opposite of what happens. As a native Danish speaker I'd write "Du kan vende tilbage til originalerne ved at holde musen over dem.". I've had a hard time finding a translation that more accurately matches the wording of the original. The best I could come up with is "Du kan reversere oversættelserne ved at holde musen over dem.", but that just sounds like you're speaking English with Danish words to me, so I don't know if it's useful.
In French the second line is translated as "Plus la difficulté augmente, plus la traduction est importante" which means "As the difficulty increases, translation becomes more important.". Kagi Translate proposes "Plus la difficulté augmente, le volume de traduction augmente" and a few other things that don't look quite right to me.
I don't know how much this matters, since you'd probably end up exposed to a lot of different translations of many different sentences with this tool. Statistically, most of those will be correct and so you'll end up good enough understanding of the other language anyway.
In any case, you'd probably want to make extra sure that the examples on the front page are absolutely correct, so I hope you find my two corrections useful.
Oh! One more thing... when you select Japanese it says "Supports Furigana", but there's no furigana shown in the example. It would've been cool to see that on the example page as well.
I hadn't heard of Kagi Translate; I'll add it to the benchmarking I do. And I'll see about adding Furigana there.
And why do you think did duolingos competitor "toucan" not take off? How does nuenki do the job better?
Seriously interested and happy to read more insights.
Disclaimer: i did create an MVP of this exactly. But it was just to try and generate some cash quickly to finance my main project (dumb idea, of course). I did get some signups but never even launched the app (and not planning to). Just curious what your experience is.
Toucan did take off; it's fairly well known, and if their website is anything to go by they have hundreds of thousands of people using it.
It translates on a per-word basis, which means that the translations are often simply wrong due to a lack of context. Nuenki doesn't translate single word "sentences" by default, despite me spending a few days trying to improve the quality, because there just isn't enough context to go off of.
They try to mitigate that by only translating certain words, mostly nouns, which limits how much you can learn from it.
They also only have three difficulty levels, while Nuenki assigns a numerical score to each sentence's difficulty and has much finer grained control.
The flip side of that is that it's free, while Nuenki needs to pay for the cost of translation.
I've got ~20 paid subscribers. People on HN seem to love it, and most of my users are from here, while capital-L language learners are hard to market to. There's a lot of AI slop in language learning, and I'm really not good at marketing!
Every day, we'd start class by the teacher saying "Salvete, discipuli!" to which we'd reply "Salve, magistra!"
The fact that all these years later I still remember some things from it shows its effectiveness I suppose.
In any case, in years since, I've used Pimsleur (for other languages), which is a similar "get actual language input rather than learning a set of language rules up front" method, and I like to think it's worked decently for me at least!
> I like to think it's worked
It works as long as you do some slow and steady work at it. I don't think it will work if you drop-in for a couple of days every few months, read something, and then disappear.
You might remember a few sentences here and there. But we want to be able to understand as well as use those sentences in the applicable context.
I don't get it why everyone seems to think that translation exercises in a foreign language learning course such as Duolingo absolutely MUST result in a comprehension-less memorization process, which must be doomed to fail sooner or later, since memorization alone might not really contribute to the capability to build new combinations of memorized words.
From my experience with Duolingo, it all depends on how a learner approaches translation exercises. If you just keep sprinting through such exercises, in a sense, mindlessly, without asking yourself how each new sentence really differs from the ones you have already seen, then yes, IMHO you are likely to fail.
However, if you keep investigating, on your own accord (for example, by using an LLM) the underlying REASONS as to why each new sentence really differs from the ones you have already seen (i.e., grammar), then no, IMHO you will indeed learn how to build new language constructs and thus use the actual language.
I think the trick is to push yourself and attempt - as soon as you can - to ignore sentence "building blocks", "missing words" and "hints" provided by Duolingo and always try to build an answer to every exercise entirely from scratch in your head. That forces your brain to understand what is really going on and create a "set of rules" for using a language as opposed to only memorizing a "set of samples" of a language.
I also don't mind the "gamification" of the learning process: it allows a learner to expect more out of himself or herself by watching it not to carelessly lose the "hearts" exercise currency, by trying to earn the "gems" bonus exercise currency, by comparing himself or herself against his or her peers through leagues and leaderboards and, the last but not least, by continuing to learn every single day because of his or her running "learning streak".
Duolingo can give you only as much as you decide to get out of yourself - as is the case with any other kind of foreign language learning course. Effortless, magical learning processes simply do not exist.
This is where comprehensible input shines.
- you start reading actual long form content from day one instead of practice sentences
- the content maintains the context across its length, letting the brain use its pattern recognition apparatus
This does not happen with the grammar translation method. You lose the context. I would compare it with RAM being swapped to disk repeatedly in a low-ram situation on your computer.
I have never studied the grammar of my mother tongue. But I can speak complex sentences rapidly because my brain managed to recognize the patterns in the language and store the sequence information somewhere.
If they expend deliberate effort on it, some people might find methods like the ones Duolingo uses somewhat useful. However, I believe if you are capable of doing that, comprehensible input might give you more bang for the buck. It has, at least for me, provided faster results and a better vocabulary than grammar translation and half-hearted attempts at CI. I felt more confident with the language after 10 days of CI-based learning than the previous six months of memorizing noun and verb forms and meanings and translating random sentences.
That said, I still do about 10 minutes of Duolingo every day, just as a kick start of my daily language-learning routine. It's also an effortless way for me to pick up a few new words on a daily basis. Somehow once I did that, I have more drive to do more comprehensive input by watching Youtube videos or reading some readers.
For example, one big thing that Duolingo's method completely misses out on is the importance of a rich communicative context. This was implicitly there in Krashen's original monitor model, but wasn't fully appreciated until closer to the turn of the century.
You NEED to consume tens of thousands of words repeatedly used in different contexts for the brain to make those automatic connections. Random sentences do not maintain the context which would have otherwise helped you figure out the possible meaning of some words in the following sentences/paras. That is one of the biggest flaws of any translation method.
It certainly makes you appreciate the unbroken oral tradition by which these enormous works of literature were passed down.
I prefer short stories. I have acquired hundreds of laghukathā collections over the last couple of years and read from them as time permits.
I am working on two Sanskrit-related things at the moment:
- a website where I am putting up proof-read stories from scanned copies of old issues of the Sanskrit Chandamama
- a "Sensible Guide to Samskritam" that will use the Baroda Critical Edition of the Valmiki Ramayana as the foundation to construct a single story told across 100-odd bite-sized chapters. This will essentially be a Sanskrit version of Lingva Latina.
The first section of the Ramayanam is the Sankshepa Ramayanam, or, the Ramayanam in summary. This gives the outline of the whole story which then is expounded in detail in the subsequent sections.
So, I am avoiding the first four sargas of the Balakanda entirely in my guide. The story starts with a description of Ayodhya and then moves on to Dasharatha and his family. I want to keep things simple and linear so that the story has momentum and readers feel like continuing the story.
I will have to find a way to incorporate all the side stories without damaging the momentum. Will probably add them as "side quests" at the appropriate juncture.
> after completing the basics
I have to disagree here. Best to jump in directly using glosses (will start with an English one. Might add a Hindi one at a future date). This fetish for basics is a big hurdle that I have personally experienced.
You will never be confident enough to start reading the Ramayana no matter how much you study the language because it is a game of vocabulary.
You need vocabulary to understand things. And the only way to acquire vocabulary is to read a lot.
Once parsed, I can look up the meaning of the word on ashtadhyayi.com or elsewhere, but not before.
Take this story (https://www.adhyeta.org.in/sa/k/samskrita-chandamama/198404/...). Assume you have a gloss available for some words. Do you really need to know the pratyayas for the nouns and verbs for you to be able to understand the story?
For this story, you don't need to know the pratyayas but you do need to know the tenses, the ktva-lyap forms, etc to fully understand what is going on. With only an incomplete knowledge of those aspects, one can sort of intuit the overall meaning but eventually would find that they had the wrong idea altogether when reading the corresponding translation in a language they know.
If you have any interest in app based review (not courses - I specifically try to work with input) I would love to get feedback on the Sanskrit experience.
I posted a bunch of comments about it in the past few days, I don’t want to take over another apps thread, but there are so many cool languages being learned here
Grammar is the analytical part of language. It is not the language itself.
Try reading Spanish texts for a few days without doing these drills and you will notice a massive increase in relative comprehension.
Obviously knowing grammer isnt going ti get you all the way there, probably not even 10 percent, but its a much much sturdier foundation than "donde esta la biblioteca", etc
The direct teaching method works but is time-consuming and generally used for languages that lead to an occupation, viz. English. The grammar translation method is a waste of time. It might satisfy your intellectual curiosity about the structure of the language but you won't be able to make yourself understood after a lifetime of study. I wonder at the sheer lunacy of dumping thousands of random sentences into your lap and translating it from one language to another.
After a year and a half of false starts, I started reading a couple of Sanskrit stories every day. Because the context is maintained across the story, your brain starts recognizing patterns in sentences. You keep reading sentences like
sarvē janāḥ kāryaṁ kurvanti
sarvē janāḥ gacchanti
sarvē janāḥ namanti
and you automatically associate sarvē (all) with janāḥ (people) without needing to know the declension of those words. This applies to the cases as well.
To be able to converse about or understand a wide variety of topics, you will eventually have to move beyond stories due to restrictions on the tense/aspect/moods you encounter as a result of the nature of the material. But that is doable.
[1] Much of India is bilingual. A substantial minority might know four or more languages due to the many mother and father tongues and heavy internal migration across the states (whose boundaries were drawn on linguistic lines post-independence)