Live data from Hacker News

Duolingo reaches $6.5B valuation on day of IPO

reuters.com

221–230 of 352 posts

Re: Duolingo reaches $6.5B valuation on day of IPO

#221
As an English speaker, DuoLingo didn’t seem that interesting to me. Then I listened to a podcast where the found Luis von Ahn explained why he started the company [1]. He grew up in Guatemala, and he noticed that learning English would essentially triple anyone’s earning potential in his country. Sure, I might never learn Chinese in DuoLingo, but if I could triple my income by doing it I bet I’d be much more motivated. English speakers that think this app is a toy aren’t the target market.

Luis also invented the captcha and used it to digitize the full archive of the New York Times. I can’t speak to the quality of the instruction in DuoLingo, but I’m willing to bet that if there’s a computer-assisted approach to language learning that works, Luis is one of the people most likely to invent it.

[1] https://www.npr.org/2020/05/22/860884062/recaptcha-and-duoli...

Re: Duolingo reaches $6.5B valuation on day of IPO

#222
post #121
post #114

Earlier quoted context omitted.

> There are very few consonants, only 5 vowels, not tone and accent is irrelevant. Except that l/r is pronounced differently than most English speakers, the vowels aren't always pronounced yet take up time ("Nan desu ka" is almost always pronounced like "Nan dess ka", for example), and syllable accent is tone rather than volume for emphasis. Other than that, it's completely easy.

You say that sarcastically, but those are all easy things you can master in a week, and they are the sort of thing you might find in the English language or even a dialect. It doesn't compare to learning Chinese or Arabic it seems to me. Just the fact that you can write out romanji and get reasonably understandable Japanese makes it easier than a significant amount of languages out there. In my opinion the only thing…

A guess as a beginning Japanese learner: Japanese has a lot of homophones because it has so few syllables. It is also almost always written without spaces. Kanji is actually pretty helpful in reading because it provides semantic meaning that disambiguates homophones as well being more terse and breaking up the writing along grammatical boundaries.

I'm approaching this from the angle of wanting to understand video-game Japanese, and when I am trying to parse some 8-bit era all-katakana chunk of text I find myself wishing it included kanji even though I can only recognize a few dozen just so the grammar is more clear.

Re: Duolingo reaches $6.5B valuation on day of IPO

#223

Earlier quoted context omitted.

The problem is that language learning is inherently about putting your brain in tricky, slightly painful situations. You have to decipher confusing sentences, handle unfamiliar words and wrestle your brain into thinking in a different language. You can't really make a fun game that does this. Therefore Duolingo makes a game that focuses on translating the foreign language to your native one. That's great for starting…

>The problem is that language learning is inherently about putting your brain in tricky, slightly painful situations. I think that's a really good description of learning in general.

Growth in general really. Stretching and building muscle, for instance, or training endurance.

Re: Duolingo reaches $6.5B valuation on day of IPO

#224
post #215

Earlier quoted context omitted.

The old FSI language courses are very effective. And free and in the public domain. FSI combines hundreds of pages of print materials with hundreds of verbal audio drills. I used it to supplement a French course in college and my professor asked me how I advanced so quickly. The course materials were doing very little for the rest of the class because they didn't force natural language thinking. FSI grammar drills ma…

If you search, you’ll find plenty of people willing to sell you some “FSI” courses for hundreds of dollars. Is there an official source for FSI courses?

It looks like https://www.fsi-language-courses.org/ is what you want.

Re: Duolingo reaches $6.5B valuation on day of IPO

#225
I've been learning Russian for a year (currently at around B1-B2). None of these new age language learning apps are actually useful, even to get your feet wet. The old is still gold when it comes to language learning. You think you're learning with these apps, but in my experience you're not. They just don't seem to put anything in long term memory well. There is spaced repetition in some of them, but not enough repetition.

Good old audio course Pimsleur is better than all of them combined. Cortina Method is best resource to learn grammar and the book is so old it looks like a smudged typewriter print. All old methods are based on learning and repeating sentences after sentences rather than learning individual words. You really end up learning like a child. Pattern recognition is where it's at.

I can't recall anything I studied on Duolingo for Russian. I can't forget anything I learned on Pimsleur. And I finished the last module in January.

Re: Duolingo reaches $6.5B valuation on day of IPO

#226
post #46

I was a cryptologic linguist for the army and went through a 16 month course of 6-7 hours of classroom instruction per day M-F, native speaking instructors, tons of immersion, and a couple hours of homework each night and left still feeling like I hardly knew the language that I was supposed to be interpreting. Unless your major in college is a specific language, you probably won't achieve any sort of fluency in your…

the problem isn't that it's hard to do. the problem is that their business model relies on their app not achieving it's portrayed purpose - learning. success means losing your customers. that's why I don't use it anymore and I'm baffled by how good they are at being bad. or rather - staying on the exact thin border between "not good enough" and "not bad enough".

Re: Duolingo reaches $6.5B valuation on day of IPO

#227
post #167

Earlier quoted context omitted.

Listening is easier than speaking only when what you listen to is easy and at a pace you can follow where you can have things repeated. Speaking is hard, but those native in the language is more likely to understand slow and broken speech with a little effort than a learner is to understand fast and unclear speech that takes shortcuts. This is my experience with learning Mandarin at least, both using classroom teachi…

> Listening is easier than speaking only when what you listen to is easy and at a pace you can follow where you can have things repeated. This is definitely not the case. Yes, listening to difficult audio is really hard to interpret. Your proficiency in listening isn't defined simply by how well you can listen to people in a perfect scenario, but also your capacity to understand the language. Reading is usually easie…

If there are multiple ways to say something, I only need to know one of them to speak clearly. For listening I need to be ready for anything.

Re: Duolingo reaches $6.5B valuation on day of IPO

#228
post #46

I was a cryptologic linguist for the army and went through a 16 month course of 6-7 hours of classroom instruction per day M-F, native speaking instructors, tons of immersion, and a couple hours of homework each night and left still feeling like I hardly knew the language that I was supposed to be interpreting. Unless your major in college is a specific language, you probably won't achieve any sort of fluency in your…

>The problem is that you can't possibly become fluent in a language with an app like it and there isn't any single app that will. I don't think anyone is seriously under the delusion that Duolingo alone will make you fluent in a language. It is an amazingly powerful tool to bootstrap you from "I cannot understand a single word of this language" to "I can pick up a newspaper and get a vague idea of what they're talkin…

> I don't think anyone is seriously under the delusion that Duolingo alone will make you fluent in a language.

I'm active in various Facebook/Reddit groups dealing with language learning. The average user, who's never learned a language before, does think DL will make them fluent. And DL advertises themselves in this manner as well. It's a common issue: "I've got my entire tree gold, why can't I understand natives?"

> I cannot think of a single more effective method than Duolingo to get over this first hurdle.

I think an actual coursebook is more useful, but it also likely depends on person. DL doesn't really teach language skills for most the courses. It has you translate, you never practice reading for comprehension, listening for comprehension (or listening to natives at all, in the courses with TTS), or writing. And not to mention the default of 'click the words' in the mobile app doesn't force recall at all.

Re: Duolingo reaches $6.5B valuation on day of IPO

#229
post #104

Earlier quoted context omitted.

I've learnt both English and German at school, for pretty much equal amounts of time - about 12 years each. And well, I can speak fluent English, but barely any German, despite usually having at least ~2 hours of lessons a week, plenty of homework, oral exercises..... For me, it has everything to do with actually using the languge in practice. English was and is everywhere - it's much easier to practice when all your…

Immersion is absolutely essential to language learning and English has an abundant amount of resources, as you say. I understand the difficulties you've had with getting immersed with German, given that my target language while in the military had very little content period, let alone things I was particularly interested in. It might be worthwhile to look for German translations of works you've read/watched in your n…

If you use Netflix and VPN into Germany or Austria to watch movies then many of them have closed captioning offered only in German.

Re: Duolingo reaches $6.5B valuation on day of IPO

#230

Honestly from using duolingo pretty heavily for a couple of years, attending language classes and doing my own individual study I've found duolingo pretty useless. It's "fun" and gives one the sense of progress without really accomplishing much.

I'm trying a language learning startup and surveyed a lot of people. The response is universal that everyone's tried Duolingo, and it didn't work for them in actually learning, despite monster streaks. I think it's masterful as a dopamine machine and employing habit building tactics. Perhaps it's a good icebreaker for people in language learning, as long as people are resourceful enough to look beyond Duolingo.

Duolingo has worked great for me. I had learned Spanish a decade ago and completely forgotten it. In a few months of consistent practice, I learned enough to get around. And after 6-9 months, I could have basic conversations.

I was never good at learning languages and Duolingo has been the best system for me. I like the paced repetition it does. Also, that it has you practice reading, writing, saying, and listening.

Post reply on HN