Hacker News Reader

Comment by dmurray | Hacker News Reader

This was the most insightful snippet in the article for me:

> Like all the teachers I spoke to, Zimotti sees Duolingo as supplemental to the kind of deep immersion that language learning requires. But, in his opinion, the time most people spend on Duolingo is time they would otherwise spend on TikTok or watching television, not learning a second language in some more optimal way.

Unfortunately, Duolingo doesn't market itself that way, preferring nonsense like "fifteen minutes a day can teach you a language". Of course there are many more customers who can spend fifteen minutes a day on it than hours at a time in an immersive context, topped up with fifteen minutes of Duolingo.

View on HN

I think it's perfectly adequate to learn A1 level of a new language. I actually did it with German and started in-person classes from level A2.

Agreed. Duolingo is great for first exposure. I learned how to read pinyin, ask short questions (e.g. where's the bathroom), make simple purchases, etc.

Beyond that, I find it almost useless for learning grammar and vocabulary. But this is still an important niche, and Duolingo owns it completely.

I got my A2 German from Duolingo and dubbed Netflix shows (German audio, English subtitles). Gonna start B1 in person in a couple of months.

But while it felt like I was starting from scratch, I didn't. I had 4 years of German in middle school. I can't tell if I just "unlocked" that part of my brain, but what I can comfortably say is that ~1 year of Duolingo+Netflix made me feel much more comfortable than a more formal learning method throughout middle school.

Fifteen minutes a day can teach you a language. In another comment on here, I mention that the FSI estimate that a group one language takes 480 hours to learn. If you study for fifteen minutes a day, that means you will need to study for 5 and 1/4 years to master your language. Duolingo’s marketing is not the issue - the issue is the majority have unrealistic expectations regarding the effort and time involved, get frustrated and then give up.

This assumes there's no economy of scale to study time, and I'm just not sure that's true. It's not indefinite, because people tire out, but I'm pretty sure an hour long study session is more than 4x as effective as a 15 minute session.

It’s normally in the pop science articles that concentration starts bonking out around 40mins so I assume there are some studies backing that up. Personally I would take 7 x 15 minute sessions per week over a single 1hr 45 minute session or an hour session and a separate 45 minute session.

The down-time lets you absorb that knowledge

Commonly repeated misconception. There is no way in h anyone learns Spanish in 480 hours. The number you are looking for is 575-600 class room hours which conveniently ignores this is 24 weeks with 7 hours of homework per week. Under nearly ideal circumstances where one's life is dedicated to the task. So starting point for consideration with self study is over 768 hours to reach a Spanish level you'll be moderately pleased with.

My view is that 800ish hours is the first point where you start to get a bit of reward for having studied the language & likely can keep your ability long term without forgetting it all. But it is more of a starting point for study if trying to get fluent/highly functional/professional.

Your numbers are not true. Or, are true only if you demand some very high level of knowledge.

I sux at learning foreign languages and did learned enough to converse and communicate everything needed with less then 480hours of learning. That was years ago, before Neflix and internet were available. And the other kids in the same school learned more then I did.

Moreover, English is quite close to Spanish, so any English speaker should learn much much faster. Many words are for free.

> Under nearly ideal circumstances where one's life is dedicated to the task.

It definitely was not and I would not even consider it ideal.

> So starting point for consideration with self study is over 768 hours to reach a Spanish level you'll be moderately pleased with.

Only if you have some kind of super great expectation.

To make sure we're on the same page, at the level you describe, were you able to confidently watch YouTube videos on non-fiction topics like documentaries and generally always understand? Or for example dramas? Just checking because being able to produce language without matching listening skills to understand everything they say back is a kind of popular YouTube fake polyglot thing these days sadly, and listening is the most unfakeable skill of them all. No technique on earth can shortcut building the robust neural connections needed to instantaneous interpret unmodified native speech.

Documentaries yes, those are the easiest content to watch. Youtube did not existed and in general getting your hands on videos was hard.

I was able to communicate with French tho. We traveled there and everyone lived in host family for over a week. There was no issue communicating with French, whether adults or peers. Most books were unpleasant to read, but I generally understood most of the normal ones. It just took too much effort and there were some paragraphs I did not understood here or there.

Also, movies are much harder to understand then real people. I did not had any issue conversing with Americans for literally years and still can't understand movies fully. Maybe if you learn from movies entirely and watch them all the time it is different, but if the goal is actual communication it is not necessary benchmark.

The issue here is that you or many people put absurdly high requirements as a benchmark for "knowing language" or "knowing enough to be useful" for the sake of an argument. Like dramas - they use weird vocabulary, phrasing and sentence structure. You can literally discuss philosophy in a language and still not understand that.

From the wikipedia on the Foreign Service Institute:

FSI provides more than 800 courses—including up to 70 foreign languages—to more than 225,000 enrollees a year from the U.S. Department of State and more than 50 other government agencies and the military service branches.[3]

I think I'm going to take their estimate over yours.

https://en.wikipedia.org/wiki/Foreign_Service_Institute

The FSI estimate is still just classroom hours. They say this.

Duolingo’s marketing is not the issue - the issue is the majority have unrealistic expectations regarding the effort and time involved, get frustrated and then give up.

Yet if DL maintains that "Fifteen minutes a day can teach you a language" (in addition to being "The world's best way to learn a language") -- then it would seem that its marketing quite definitely is at fault.

Being as it is (in part) from ad copy like this where these misconceptions originate.

I love duolingo, but you wont master language with duolingo alone no matter for how long. There is a ceiling of how much it teaches. And to their credit, their own claim is that they teach up to B2 level in reading and listening.

> Unfortunately, Duolingo doesn't market itself that way, preferring nonsense like "fifteen minutes a day can teach you a language"

tbf, the ad says "fifteen minutes a day can teach you a language, what can fifteen minutes of social media do?"

Help you keep in touch with friends? You may even have some of those yourself. The 1-to-many nature of the medium may be more efficient, and allowing for pictures (which are worth a thousand words) is also helpful.

4 hours of social media a day, on the other hand, might be excessive.

Reply on news.ycombinator.com