Been trying this over the last week. It's helped me enjoy watching shows entirely in Japanese while also making the sentence mining for Anki so much less tedious. Thanks for your good work.
Kinda sad Anki was not mentioned once, I'm really invested in it for japanese and medschool. I don't see myself changing to another SRS sadly. Anyway good job on improving the Migaku system, hope the best for you guys.
We still support and plan to continue supporting Anki. We just don't think it's a tool that most new learners can easily connect with so we won't be making it a showcase piece of our videos. Anki definitely serves power users well so we will continue supporting it. 👍
@@MigakuOfficial I think I'm going to start trying Migaku memory. But I have a question. If my subscription runs out, can I export my cards from it? Is it better to back up my cards to Anki and use both simultaneously just in case?
@@speedytruck We will add some ability to export data out of Migaku Memory at some point in the future but currently a subscription is required to use it.
this sounds so nice but the problem is that you realize how significant missing one word is when talking with someone, especially when those words outside of the 80% are uncommon and usually very specific, so often you completely miss the entire meaning because you didn't get one of the keywords in the sentence. now imagine how many words are spoken in a 2-minute conversation and start to think of all the words you didn't understand. language normally requires 100% knowledge for understanding, especially when it's a foreign language where context is more difficult to read because you haven't been speaking it your whole life. Not trying to knock what you're saying, regardless you have to study those 80%, but it's misleading to make it seem like you can almost stop there or once you're there you're just good to go. You have to keep going or you will still find yourself wondering wtf those Japanese people just said to you (this does not even take speed/listening comprehension ability into account, which is a massive part of understanding)
No where in the video does it imply that you don't need to continue learning more vocabulary. Rather this video is about the approach to reaching a point where you can understand all words in a sentence but one or at most two, this is a great place to be as it allows acquiring new words in the context of a sentence where the word you don't know is the only unknown word in the sentence. This premise is clearly stated at the beginning of the video. Also a frequency approach still makes the most sense even though yes, to reach the highest levels of vocabulary a ton of vocabulary must ultimately be learned. We've replied to similar comments that a couple other users posted about this topic, so if you'd like to hear more about this from us then please reference our replies there.
@MigakuOfficial I never said that you said you don't need to learn more, but the way present learning the 80% is like wow 80% is so much that you'll be able to understand almost everything, which you reinforced in your reply by saying "all words in a sentence but one or at most two" and the point I was trying to make as a person who has been learning Japanese for a year and half and knows probably right around 1500 words currently is that missing one or two words in a sentence is astronomical to your ability to understand most of the time. Context is much harder to use in a foreign language, especially when the words you don't understand are critical to the meaning of the sentence, unlike when hearing an English sentence and some word is used that has similar roots to other words and the context of the sentences is much more familiar. I was not discounting your video, I am simply saying that it's a TAD BIT misleading because it makes it seem like missing a word or two is not a big deal because you will still be able to understand or learn from the context what those missed words mean and I'm saying that's not really true and almost never works out that way, from my own experience. Maybe I'm unique and my brain works differently than everyone else, but it's my thoughts on the subject.
@@DarkKnight4533 The key of learning by frequency is that first learning the top 80% of words starts to unlock a lot of content by making lots of sentences only have a single unknown word in them. These sentences then can be usually be understood when looking up the unknown word and thus makes learning new vocabulary easier to learn. Then continuing to focus on learning by frequency until around 90% coverage is a great strategy to unlock many many 1 Target sentences in this way, and it makes the language really approachable and easy to learn from using a tool like Migaku. This video is about the approach of learning vocabulary, and yes, a frequency approach allows the fastest growth in understanding objectively.
Supposedly, high school students know around 36 000 words of their native tongue passively after they leave high school. Putting the average native speaker in the 99 percentile. Therefore, to become fluent, one must also strive to acquire the same amount eventually. It would probably take 5+ years through daily input.
This 80-20 is based on a false premise that each word has the same value as any other word. They don't. Some words carry more meaning than the others in linguistics. Functional words contribute to very little of the meaning of a giving sentence (though essential) and some words carry most of the meaning because they are WHAT we are talking about. So, in a sentence with 12 words you may understand only 5 and get a general sense of the sentence. But in another sentence of 12 words you may get 10 of the words and still have no IDEIA what people are talking about. To be really functional in a language you're gonna need 30,00 words and above. To really start playing around you're gonna need at least 10,000. I've learned Japanese (plus English, Italian, Vietnamese and now I'm keeping improving my Vietnamese and just started Mandarin). When learning Japanese I believed in this BS. The core 6k WILL NOT make you able to understand most things in Japanese. It's a delusion. Things start to get interesting only after the 10,000 words. ANY conversation between two educated adults will have WAY MORE than 6,000 words.
Another misconception is that people don't understand how many words appear in a giving amount of time so they think the "rare" words would occur once in a lifetime. Not at all! A person speaking English on average speaks 110-150 words per minute. In 1 hour we get between 7,500 to 9,000 words being spoken! That's A LOT of opportunities for "rare' words to be used. In 1 hour a 'more rare' word can be used a good couple of times. "rare" here means only in comparison to the other most common word. Not in absolute. You'll see a LOT of "rare" words each hour of immersion, for instance. Just that the common words will appear way more.
@@JohnnyLynnLee I think you quite fundamentally misunderstand the basis by which we are recommending a frequency based approach. Learning the most frequent words first increases the chances of coming across new sentences with only one or two unknown words greatly. This in effect makes it much easier to understand these sentences with the aide of a tool like a popup dictionary, and thus makes learning even more new vocabulary easier. As we quite clearly state in the video, an exponentially larger number of words is required to increase the total percentage of content that can be understood. This is why we recommend focusing on specific genres of content once they get to around 90% of words understood. Genres of content share vocabulary much more closely than content outside of those genres, and thus focusing on vocabulary that commonly occurs within a genre before moving on to other genres allows a learner to see progress in understanding that specific genre more quickly than if they followed a more random approach to learning. Also the core 6k deck you are talking about was not based on frequency in the same way our course is, many of the words in it are much rarer and not frequently used.
@@MigakuOfficial I AGREE learning based on frequency is useful. What I don't agree anymore, because it's objectively false, is that with 6,000 words alone (or even only 5,000 now with Migaku) you can understand "most of the content". A thing to have in mind here is that COVERAGE OF WORDS does not equal COVERAGE OF MEANING. Because, as I've said, it's not regarding meaning each word has the value of 1. Some words are what's is called "meaning deep", they have, say, value of 10, some has value of, say, 3 or 4. Another thing that is important to understand here is that a "rare" words isn't rare IN ABSOLUTE, it's rare in CMPARISON. with a person talking for one hour straight speaking around 7,000 to 9,000 words a moderately "rare" word (here I don't have the numbers, I could get later, so I'll just give an example) may appear 30 times in one hour among 9,000 words! It's a lot of appearances. Only that the most common will appear 800 times. So "rare" words appear OFTEN in watching one hour of someone speaking or reading 30-60 pages of a book in 1 hour. It's like learning Chinese after learning Japanese like me. "Only a minority of hanzis were simplified. True. But a minority of around 8,000 hanzis is still around 2,00 hanzis/kanjis! And generally they are the most common. So you'll see them every time even if BY COMPARISON they are a minority of hanzis. That's just what I want people to have in mind so they don't become frustrated at the end, like I did.
Graded readers. Graded readers are the key. You get this, automatically - and you get it in context, and you get it with grammar. I highly highly highly recommend the Japanese Language Park series for beginners.
Hi Yoga, do you think you will do a word frequency list using Netflix for other languages, like Spanish. What you've got there is pure gold. Great job!
Looks wonderful! Do you have the frequency list also for Chinese? Is the Chinese learning experience in Migaku similar to the Japanese? Will the planned OCR module help to mine subtitles from Chinese streaming websites as well?
😂 unless you apply for a university level linguistics professor position, you rarely (if ever) will need the proficiency level above JLPT N3 / B2 / IELTS 6 (at least grammar wise)
When will you update the Anki addons? I was a huge Migaku supporter back when you were doing the addons (I even stuck around after the Matt vs Japan drama)... I know they don't make you money, but please update them! For the community!
We have an up to date Anki addon called the Migaku Anki Addon that combines with out Migaku browser extension to provide the same functionality as our previous addons, please check it out. The other addons are discontinued.
@@huginnmunnin5998 beginner. kun-yomi is the native japanese word and that's necessary for speaking Japanese. on-yomi are the foreign Chinese words and it's more important to be able to read the meaning of the kanji then the Chinese pronunciation. Even if you don't know the on-yomi, you can still read it as "old mountain dragon" but you won't know the pronunciation.
@@Adam-jr4lx Yeah I know. Just wanted to understand your process and how successful it can be. I’ve been studying jp for 7 months, and when it comes to kanji, I’ve seen that is usually better to try to memorize only the meaning first, then learn 5 or so vocabulary words that use that given kanji. So far I’ve memorized about 800 kanji in this timeframe and I my brain got used to learning it this way. 頑張ってね！
It's currently in internal testing and we hope to release it to Early Access in the next two weeks, and then a week or two after it will be released to all users.
A key thing to remember (as its often not stated) is that acquiring 80% of the most frequently used words does NOT equal 80% understanding generally. Take these 2 example sentences; I went to the aquarium last week with my friend from work. I went to the strip club last week with my friend from work. If 'aquarium' or 'strip club' were the only words you didn't know in your target language you would be well over the 80% mark in terms of vocab you understood, but that does NOT mean you got 80% of the meaning... 'I went to the *blank* last week with my friend from work' is almost a meaningless sentence when you think about what the actual point would be in someone trying to communicate this. You went to 'a place' last week? Cool story bro! Unfortunately the entire point of most sentences is tied up in words that by definition are less frequently used! That's why they are the interesting parts! Being able to understand that 'someone went to some place last week' is great.... its a start.... but you are still very far from true communication. I don't say this to be a downer.... but as this often goes unsaid I think it leaves people thinking they can just cram in say 20 words a day for 3 months and 'understand 80% of the language'. You cant. You can understand 80% of the individual words that make up sentences. That's a very different thing.
As is stated in the video, your first step is to be able to understand what you don't understand in a sentence. So yes, a frequency approach is the best way to get to the point where you can pick out the words you don't understand. With the help of tools like a popup dictionary it is trivial to look up the words in sentences like you've mentioned and then often gain a full understanding of them. As acknowledged and emphasized in the video, an exponentially larger number of vocabulary is required to reach the highest levels of raw understanding. Which is why focusing on genres of interest after reaching something like a 90% general coverage of words is recommended, as much less vocabulary is used within specific genres of content on average than in contents of all genres of content. That being said, a frequency approach is the best way to approach increasing your understanding of Japanese (or any other language).
@@MigakuOfficial I don't fully agree or disagree really. There is nothing wrong with this approach generally and talking about the 80-20 rule per se. I just think its sometimes used as a bit of a false promise. Although not stated directly, learners will take from this that 80% of vobab will = 80% comprehension. But lets say it takes someone a year to nail down everything they can from the 80% list. What happens when words like 'Aquarium' fall squarely outside that list, or even the 85% list and you're advising beginners not to look that stuff up too early? Do they just miss the entire point of these sentences for that full year? What happens when MOST sentences are like that in specific areas of their immersion? Do they just suffer that feeling of not quite knowing fully what's going on? might that not lead to problems with motivations later?
@@mowgli123456789 They look up the word "aquarium" in that case and then likely gain a full understanding of the sentence it is being used in, and if they see the word continuously come up in the content they are frequently watching and are having trouble understanding it still then they can create a flashcard for that word for extra practice. The first step of unlocking a language is getting it to the point where you're only not understanding a word or two in each sentence, from there you enter the vocabulary acquisition phase which basically lasts as long as a learner wants it too. But yes, language learning is a lot of work and to reach the highest levels of understanding, a learner must learn many many words. That all being taken into account, a frequency approach is definitely the best approach to tackling a language. Remember that you can still look up and organically begin learning the words that fall outside of the 90% frequency range, but that actually creating flashcards and focusing on words that fall within that 90% range will lead to the greatest raw understanding gain of the language as a whole. At a certain point though targeting understanding the language as a whole becomes a losing strategy and focusing on genre specific vocabulary is necessary to continue making tangible progress towards understanding that specific genre well though.
@@MigakuOfficial For context Ive been doing flashcard reps daily for about 17 years and i use this method exclusively - I create cards where I know every word except the one Im learning to help the learning process via context. In thousands of cards I have maybe 10 that are single words with no contextual sentence, So Im fully aware of the logic behind what you are saying. When you said you advise against learning things that are rarer than the top 95% of words (early on) I took that to mean you were against even looking them up as if it was a waste of time at that level. That's basically where I disagreed. If that's not the case are you are merely saying its probably not worth added them into your flashcard system yet, then yeah, i don't really see the problem with that. Look it up (if its important to you to know what's going on right this second) then allow yourself to forget it as you likely wont need it again for some times anyway. Free yourself up to focus on more common words that will then aid in future learning via added context. Makes sense. I still personally add in any old thing I come across these days... but I guess Im so deep in now it doesnt matter lol.
@@mowgli123456789 Yeah, you can still look up the words of course, but you should just not focus on studying them explicitly is all. And yes, once one gets really far into a language they can basically create cards for anything they personally want to learn as they will by that point have a pretty good intuitive understanding for the scenarios specific words will come up in by that point. 👍
This 80-20 is based on a false premise that each word has the same value as any other word. They don't. Some words carry more meaning than the others in linguistics. Functional words contribute to very little of the meaning of a giving sentence (though essential) and some words carry most of the meaning because they are WHAT we are talking about. So, in a sentence with 12 words you may understand only 5 and get a general sense of the sentence. But in another sentence of 12 words you may get 10 of the words and still have no IDEIA what people are talking about. To be really functional in a language you're gonna need 30,00 words and above. To really start playing around you're gonna need at least 10,000. I've learned Japanese (plus English, Italian, Vietnamese and now I'm keeping improving my Vietnamese and just started Mandarin). When learning Japanese I believed in this BS. The core 6k WILL NOT make you able to understand most things in Japanese. It's a delusion. Things start to get interesting only after the 10,000 words. ANY conversation between two educated adults will have WAY MORE than 6,000 words.
Another misconception is that people don't understand how many words appear in a giving amount of time so they think the "rare" words would occur once in a lifetime. Not at all! A person speaking English on average speaks 110-150 words per minute. In 1 hour we get between 7,500 to 9,000 words being spoken! That's A LOT of opportunities for "rare' words to be used. In 1 hour a 'more rare' word can be used a good couple of times. "rare" here means only in comparison to the other most common word. Not in absolute. You'll see a LOT of "rare" words each hour of immersion, for instance. Just that the common words will appear way more.
I think you quite fundamentally misunderstand the basis by which we are recommending a frequency based approach. Learning the most frequent words first increases the chances of coming across new sentences with only one or two unknown words greatly. This in effect makes it much easier to understand these sentences with the aide of a tool like a popup dictionary, and thus makes learning even more new vocabulary easier. As we quite clearly state in the video, an exponentially larger number of words is required to increase the total percentage of content that can be understood. This is why we recommend focusing on specific genres of content once they get to around 90% of words understood. Genres of content share vocabulary much more closely than content outside of those genres, and thus focusing on vocabulary that commonly occurs within a genre before moving on to other genres allows a learner to see progress in understanding that specific genre more quickly than if they followed a more random approach to learning. Also the core 6k deck you are talking about was not based on frequency in the same way our course is, many of the words in it are much rarer and not frequently used.
@@JohnnyLynnLee You are either confused or in your ego. I think it's both. Do you really think that the first word one should learn is a word like "Merulaxis stresemanni" instead of words like "eat ", "sleep" , "go" ?!? Are you serious or are you trolling? Please tell us that you were trolling and joking around.
@eliiste I repeatedly stated here, and by now even a 5 year old child with severe cognitive deficiency should understand, honestly, that I'm NOT against learning by frequency and I always try to do it myself. What I am saying instead is that 6,000 words will NOT get you even NEAR to understand most shows, even simple ones, and entertain ANY type of meaningful conversation with ANY educated adult. And that's a FACT. And about frequency order, although, as I'm saying here by the zillionth time, I'm IN FAVOR or using frequency, many successful learners that are proficient in DOZENS of languages, like Steve Kaufman don't care AT ALL about that. And his argument is true too. He says you shouldn't mind the most common words because they will appear ANYWAY. So although I DO use frequency lists and stuff I tend to right away ignore the first 500 or so, because I'll learn them anyway reading right in the first months in a language.
@eliiste In other words I'm NOT saying you shouldn't study the most common 6,000 words. What I'm saying is that you FIRST ANKI deck in a language should have instead AT LEAST the 10,000 most common words or above. So you can START to play with the language. To really understand it you'll need the first 30,000 to 60,000 most common words. It's the NUMBERS I'm talking abut. Not they being in A frequency order. That's good. And note I said "A' frequency order not "the" frequency order. Because there's that too. Depending on your database the order will vary WILDELY! The MIGAKU deck, for instance, is a frequency order based on NETFLIX subs. That will have, of course, a lot of overlap, but also a lot o differences (specially in the ranking the word appear, not so much the presence or not of a word) if they were to use shows like the Abema Prime (which in my opinion would be way better) or light novels or REAL BOOKS, no fiction real books, or the news, or whatever. If you run the same program as they do you'll see words that appear right on top on one list moved WAY DOWN on the list in others, sometimes by hundreds or even THOUSANDS of positions in the first 10,000 words.
Well, just reduce the account of repetition of the 80/20 and word counts. I heard the 80%, 85%, 90% etc too many times. Please don't get me wrong, it was an informative video but I suspect the drop off rate is high in your stats. Thank you.
@@MigakuOfficial I don't think the video is too long. It's perfect. It explained the importance of focusing on the most frequent words. All explanations were tied to the process of how Migaku recommends us to study certain words. It's almost a tutorial which I like... It was a a well-done video. If somebody dropped out just because they heard 80%,85 or 90% "too many times", they certainly don't have the grit necessary to learn a language.Your thoughts and statements exposed you @iosmusicman
Bruh did you even watch the video? It’s 80 percent of the words used in Japanese Netflix. which movies and tv shows obviously use different vocab then real life, but this is a very good list to understand tv shows and movies in immersion
Yes, also the key of learning by frequency is that first learning the top 80% of words starts to unlock a lot of content by making lots of sentences only have a single unknown word in them. These sentences then can be usually be understood when looking up the unknown word. Then you can focus on common words first again and continue to increase your vocabulary unlocking even more sentences. So yes, this video is about the approach of learning new vocabulary and is not emphasizing how much you will understand at 80% (or 85% or 90% for that matter).
@@orlando5366 This is what 80% comprehension looks like: “Bingle for help!” you shout. “This loopity is dying!” You put your fingers on her neck. Nothing. Her flid is not weafling. You take out your joople and bingle 119, the emergency number in Japan. There’s no answer! Then you muchy that you have a new befourn assengle. It’s from your gutring, Evie. She hunwres at Tokyo University. You play the assengle. “…if you get this…” Evie says. “…I can’t vickarn now… the important passit is…” Suddenly, she looks around, dingle. “Oh no, they’re here! Cripett… the frib! Wasple them ON THE FRIB!…” BEEP! the assengle parantles. Then you gratoon something behind you…
Been trying this over the last week. It's helped me enjoy watching shows entirely in Japanese while also making the sentence mining for Anki so much less tedious. Thanks for your good work.
Welcome back, Yoga.
Looking forward to similar guides for Mandarin as well.
Kinda sad Anki was not mentioned once, I'm really invested in it for japanese and medschool. I don't see myself changing to another SRS sadly.
Anyway good job on improving the Migaku system, hope the best for you guys.
We still support and plan to continue supporting Anki. We just don't think it's a tool that most new learners can easily connect with so we won't be making it a showcase piece of our videos. Anki definitely serves power users well so we will continue supporting it. 👍
@@MigakuOfficial I think I'm going to start trying Migaku memory. But I have a question. If my subscription runs out, can I export my cards from it? Is it better to back up my cards to Anki and use both simultaneously just in case?
@@speedytruck We will add some ability to export data out of Migaku Memory at some point in the future but currently a subscription is required to use it.
These guys have changed my life!
Wow nicely done mining the data. Truly puts it into context.
Thanks for the frequency list! Very helpful
❤ a very smart, sane and realistic approach!
this sounds so nice but the problem is that you realize how significant missing one word is when talking with someone, especially when those words outside of the 80% are uncommon and usually very specific, so often you completely miss the entire meaning because you didn't get one of the keywords in the sentence. now imagine how many words are spoken in a 2-minute conversation and start to think of all the words you didn't understand. language normally requires 100% knowledge for understanding, especially when it's a foreign language where context is more difficult to read because you haven't been speaking it your whole life. Not trying to knock what you're saying, regardless you have to study those 80%, but it's misleading to make it seem like you can almost stop there or once you're there you're just good to go. You have to keep going or you will still find yourself wondering wtf those Japanese people just said to you (this does not even take speed/listening comprehension ability into account, which is a massive part of understanding)
No where in the video does it imply that you don't need to continue learning more vocabulary. Rather this video is about the approach to reaching a point where you can understand all words in a sentence but one or at most two, this is a great place to be as it allows acquiring new words in the context of a sentence where the word you don't know is the only unknown word in the sentence. This premise is clearly stated at the beginning of the video. Also a frequency approach still makes the most sense even though yes, to reach the highest levels of vocabulary a ton of vocabulary must ultimately be learned. We've replied to similar comments that a couple other users posted about this topic, so if you'd like to hear more about this from us then please reference our replies there.
@MigakuOfficial I never said that you said you don't need to learn more, but the way present learning the 80% is like wow 80% is so much that you'll be able to understand almost everything, which you reinforced in your reply by saying "all words in a sentence but one or at most two" and the point I was trying to make as a person who has been learning Japanese for a year and half and knows probably right around 1500 words currently is that missing one or two words in a sentence is astronomical to your ability to understand most of the time. Context is much harder to use in a foreign language, especially when the words you don't understand are critical to the meaning of the sentence, unlike when hearing an English sentence and some word is used that has similar roots to other words and the context of the sentences is much more familiar. I was not discounting your video, I am simply saying that it's a TAD BIT misleading because it makes it seem like missing a word or two is not a big deal because you will still be able to understand or learn from the context what those missed words mean and I'm saying that's not really true and almost never works out that way, from my own experience. Maybe I'm unique and my brain works differently than everyone else, but it's my thoughts on the subject.
@@DarkKnight4533 The key of learning by frequency is that first learning the top 80% of words starts to unlock a lot of content by making lots of sentences only have a single unknown word in them. These sentences then can be usually be understood when looking up the unknown word and thus makes learning new vocabulary easier to learn. Then continuing to focus on learning by frequency until around 90% coverage is a great strategy to unlock many many 1 Target sentences in this way, and it makes the language really approachable and easy to learn from using a tool like Migaku. This video is about the approach of learning vocabulary, and yes, a frequency approach allows the fastest growth in understanding objectively.
Any update on the Japanese guide being released ?
Supposedly, high school students know around 36 000 words of their native tongue passively after they leave high school. Putting the average native speaker in the 99 percentile. Therefore, to become fluent, one must also strive to acquire the same amount eventually. It would probably take 5+ years through daily input.
Yeah, the highest levels of fluency will eventually require learning upwards of 30,000 words.
This 80-20 is based on a false premise that each word has the same value as any other word. They don't. Some words carry more meaning than the others in linguistics. Functional words contribute to very little of the meaning of a giving sentence (though essential) and some words carry most of the meaning because they are WHAT we are talking about. So, in a sentence with 12 words you may understand only 5 and get a general sense of the sentence. But in another sentence of 12 words you may get 10 of the words and still have no IDEIA what people are talking about. To be really functional in a language you're gonna need 30,00 words and above. To really start playing around you're gonna need at least 10,000. I've learned Japanese (plus English, Italian, Vietnamese and now I'm keeping improving my Vietnamese and just started Mandarin). When learning Japanese I believed in this BS. The core 6k WILL NOT make you able to understand most things in Japanese. It's a delusion. Things start to get interesting only after the 10,000 words. ANY conversation between two educated adults will have WAY MORE than 6,000 words.
Another misconception is that people don't understand how many words appear in a giving amount of time so they think the "rare" words would occur once in a lifetime. Not at all! A person speaking English on average speaks 110-150 words per minute. In 1 hour we get between 7,500 to 9,000 words being spoken! That's A LOT of opportunities for "rare' words to be used. In 1 hour a 'more rare' word can be used a good couple of times. "rare" here means only in comparison to the other most common word. Not in absolute. You'll see a LOT of "rare" words each hour of immersion, for instance. Just that the common words will appear way more.
@@JohnnyLynnLee I think you quite fundamentally misunderstand the basis by which we are recommending a frequency based approach. Learning the most frequent words first increases the chances of coming across new sentences with only one or two unknown words greatly. This in effect makes it much easier to understand these sentences with the aide of a tool like a popup dictionary, and thus makes learning even more new vocabulary easier. As we quite clearly state in the video, an exponentially larger number of words is required to increase the total percentage of content that can be understood. This is why we recommend focusing on specific genres of content once they get to around 90% of words understood. Genres of content share vocabulary much more closely than content outside of those genres, and thus focusing on vocabulary that commonly occurs within a genre before moving on to other genres allows a learner to see progress in understanding that specific genre more quickly than if they followed a more random approach to learning. Also the core 6k deck you are talking about was not based on frequency in the same way our course is, many of the words in it are much rarer and not frequently used.
@@MigakuOfficial I AGREE learning based on frequency is useful. What I don't agree anymore, because it's objectively false, is that with 6,000 words alone (or even only 5,000 now with Migaku) you can understand "most of the content". A thing to have in mind here is that COVERAGE OF WORDS does not equal COVERAGE OF MEANING. Because, as I've said, it's not regarding meaning each word has the value of 1. Some words are what's is called "meaning deep", they have, say, value of 10, some has value of, say, 3 or 4. Another thing that is important to understand here is that a "rare" words isn't rare IN ABSOLUTE, it's rare in CMPARISON. with a person talking for one hour straight speaking around 7,000 to 9,000 words a moderately "rare" word (here I don't have the numbers, I could get later, so I'll just give an example) may appear 30 times in one hour among 9,000 words! It's a lot of appearances. Only that the most common will appear 800 times. So "rare" words appear OFTEN in watching one hour of someone speaking or reading 30-60 pages of a book in 1 hour.
It's like learning Chinese after learning Japanese like me. "Only a minority of hanzis were simplified. True. But a minority of around 8,000 hanzis is still around 2,00 hanzis/kanjis! And generally they are the most common. So you'll see them every time even if BY COMPARISON they are a minority of hanzis.
That's just what I want people to have in mind so they don't become frustrated at the end, like I did.
Graded readers. Graded readers are the key. You get this, automatically - and you get it in context, and you get it with grammar. I highly highly highly recommend the Japanese Language Park series for beginners.
Yeah, graded readers are cool.
@LionKimbro Thank you for the recommendation!
Hi Yoga, do you think you will do a word frequency list using Netflix for other languages, like Spanish. What you've got there is pure gold. Great job!
We would like to do them eventually yes!
Looks wonderful! Do you have the frequency list also for Chinese? Is the Chinese learning experience in Migaku similar to the Japanese? Will the planned OCR module help to mine subtitles from Chinese streaming websites as well?
I have the same question
😂 unless you apply for a university level linguistics professor position, you rarely (if ever) will need the proficiency level above JLPT N3 / B2 / IELTS 6 (at least grammar wise)
Wow, so awesome! Lucas you looking THICC!! bet you benching 220's at this point!
Great content!
Any timeline on the French vocab deck releasing?
The French Academy Course we can't yet say when is coming, but a quick-start course for French will come this year for sure.
When will you update the Anki addons? I was a huge Migaku supporter back when you were doing the addons (I even stuck around after the Matt vs Japan drama)... I know they don't make you money, but please update them! For the community!
We have an up to date Anki addon called the Migaku Anki Addon that combines with out Migaku browser extension to provide the same functionality as our previous addons, please check it out. The other addons are discontinued.
@@MigakuOfficial Ok, thank you for your reply! I hope some more focus is brought back to Anki at somepoint!
I think the best approach is to learn the kun-yomi and not care about the on-yomi.
How experienced are you with japanese?
@@huginnmunnin5998 beginner. kun-yomi is the native japanese word and that's necessary for speaking Japanese. on-yomi are the foreign Chinese words and it's more important to be able to read the meaning of the kanji then the Chinese pronunciation. Even if you don't know the on-yomi, you can still read it as "old mountain dragon" but you won't know the pronunciation.
@@Adam-jr4lx Yeah I know. Just wanted to understand your process and how successful it can be. I’ve been studying jp for 7 months, and when it comes to kanji, I’ve seen that is usually better to try to memorize only the meaning first, then learn 5 or so vocabulary words that use that given kanji. So far I’ve memorized about 800 kanji in this timeframe and I my brain got used to learning it this way. 頑張ってね！
When can we expect the course with the top 80% words and grammar lessons to be available?
It's currently in internal testing and we hope to release it to Early Access in the next two weeks, and then a week or two after it will be released to all users.
@@MigakuOfficial Any update?
When will the japanese course be out?
It will be released within the next few weeks. Just working out a few last minute technical issues.
Can't wait
I was going crazy trying to find it 😅
Looking for a guide like this one but for german
We will definitely make one in the future!
grats on the weight loss
10
A key thing to remember (as its often not stated) is that acquiring 80% of the most frequently used words does NOT equal 80% understanding generally.
Take these 2 example sentences;
I went to the aquarium last week with my friend from work.
I went to the strip club last week with my friend from work.
If 'aquarium' or 'strip club' were the only words you didn't know in your target language you would be well over the 80% mark in terms of vocab you understood, but that does NOT mean you got 80% of the meaning...
'I went to the *blank* last week with my friend from work' is almost a meaningless sentence when you think about what the actual point would be in someone trying to communicate this. You went to 'a place' last week? Cool story bro!
Unfortunately the entire point of most sentences is tied up in words that by definition are less frequently used! That's why they are the interesting parts! Being able to understand that 'someone went to some place last week' is great.... its a start.... but you are still very far from true communication.
I don't say this to be a downer.... but as this often goes unsaid I think it leaves people thinking they can just cram in say 20 words a day for 3 months and 'understand 80% of the language'. You cant. You can understand 80% of the individual words that make up sentences. That's a very different thing.
As is stated in the video, your first step is to be able to understand what you don't understand in a sentence. So yes, a frequency approach is the best way to get to the point where you can pick out the words you don't understand. With the help of tools like a popup dictionary it is trivial to look up the words in sentences like you've mentioned and then often gain a full understanding of them. As acknowledged and emphasized in the video, an exponentially larger number of vocabulary is required to reach the highest levels of raw understanding. Which is why focusing on genres of interest after reaching something like a 90% general coverage of words is recommended, as much less vocabulary is used within specific genres of content on average than in contents of all genres of content. That being said, a frequency approach is the best way to approach increasing your understanding of Japanese (or any other language).
@@MigakuOfficial I don't fully agree or disagree really. There is nothing wrong with this approach generally and talking about the 80-20 rule per se. I just think its sometimes used as a bit of a false promise. Although not stated directly, learners will take from this that 80% of vobab will = 80% comprehension.
But lets say it takes someone a year to nail down everything they can from the 80% list. What happens when words like 'Aquarium' fall squarely outside that list, or even the 85% list and you're advising beginners not to look that stuff up too early? Do they just miss the entire point of these sentences for that full year? What happens when MOST sentences are like that in specific areas of their immersion? Do they just suffer that feeling of not quite knowing fully what's going on? might that not lead to problems with motivations later?
@@mowgli123456789 They look up the word "aquarium" in that case and then likely gain a full understanding of the sentence it is being used in, and if they see the word continuously come up in the content they are frequently watching and are having trouble understanding it still then they can create a flashcard for that word for extra practice. The first step of unlocking a language is getting it to the point where you're only not understanding a word or two in each sentence, from there you enter the vocabulary acquisition phase which basically lasts as long as a learner wants it too. But yes, language learning is a lot of work and to reach the highest levels of understanding, a learner must learn many many words. That all being taken into account, a frequency approach is definitely the best approach to tackling a language.
Remember that you can still look up and organically begin learning the words that fall outside of the 90% frequency range, but that actually creating flashcards and focusing on words that fall within that 90% range will lead to the greatest raw understanding gain of the language as a whole. At a certain point though targeting understanding the language as a whole becomes a losing strategy and focusing on genre specific vocabulary is necessary to continue making tangible progress towards understanding that specific genre well though.
@@MigakuOfficial For context Ive been doing flashcard reps daily for about 17 years and i use this method exclusively - I create cards where I know every word except the one Im learning to help the learning process via context. In thousands of cards I have maybe 10 that are single words with no contextual sentence, So Im fully aware of the logic behind what you are saying.
When you said you advise against learning things that are rarer than the top 95% of words (early on) I took that to mean you were against even looking them up as if it was a waste of time at that level. That's basically where I disagreed. If that's not the case are you are merely saying its probably not worth added them into your flashcard system yet, then yeah, i don't really see the problem with that. Look it up (if its important to you to know what's going on right this second) then allow yourself to forget it as you likely wont need it again for some times anyway. Free yourself up to focus on more common words that will then aid in future learning via added context. Makes sense.
I still personally add in any old thing I come across these days... but I guess Im so deep in now it doesnt matter lol.
@@mowgli123456789 Yeah, you can still look up the words of course, but you should just not focus on studying them explicitly is all. And yes, once one gets really far into a language they can basically create cards for anything they personally want to learn as they will by that point have a pretty good intuitive understanding for the scenarios specific words will come up in by that point. 👍
Marry me bro
First, lol
This 80-20 is based on a false premise that each word has the same value as any other word. They don't. Some words carry more meaning than the others in linguistics. Functional words contribute to very little of the meaning of a giving sentence (though essential) and some words carry most of the meaning because they are WHAT we are talking about. So, in a sentence with 12 words you may understand only 5 and get a general sense of the sentence. But in another sentence of 12 words you may get 10 of the words and still have no IDEIA what people are talking about. To be really functional in a language you're gonna need 30,00 words and above. To really start playing around you're gonna need at least 10,000. I've learned Japanese (plus English, Italian, Vietnamese and now I'm keeping improving my Vietnamese and just started Mandarin). When learning Japanese I believed in this BS. The core 6k WILL NOT make you able to understand most things in Japanese. It's a delusion. Things start to get interesting only after the 10,000 words. ANY conversation between two educated adults will have WAY MORE than 6,000 words.
Another misconception is that people don't understand how many words appear in a giving amount of time so they think the "rare" words would occur once in a lifetime. Not at all! A person speaking English on average speaks 110-150 words per minute. In 1 hour we get between 7,500 to 9,000 words being spoken! That's A LOT of opportunities for "rare' words to be used. In 1 hour a 'more rare' word can be used a good couple of times. "rare" here means only in comparison to the other most common word. Not in absolute. You'll see a LOT of "rare" words each hour of immersion, for instance. Just that the common words will appear way more.
I think you quite fundamentally misunderstand the basis by which we are recommending a frequency based approach. Learning the most frequent words first increases the chances of coming across new sentences with only one or two unknown words greatly. This in effect makes it much easier to understand these sentences with the aide of a tool like a popup dictionary, and thus makes learning even more new vocabulary easier. As we quite clearly state in the video, an exponentially larger number of words is required to increase the total percentage of content that can be understood. This is why we recommend focusing on specific genres of content once they get to around 90% of words understood. Genres of content share vocabulary much more closely than content outside of those genres, and thus focusing on vocabulary that commonly occurs within a genre before moving on to other genres allows a learner to see progress in understanding that specific genre more quickly than if they followed a more random approach to learning. Also the core 6k deck you are talking about was not based on frequency in the same way our course is, many of the words in it are much rarer and not frequently used.
@@JohnnyLynnLee You are either confused or in your ego. I think it's both. Do you really think that the first word one should learn is a word like "Merulaxis stresemanni" instead of words like "eat ", "sleep" , "go" ?!?
Are you serious or are you trolling? Please tell us that you were trolling and joking around.
@eliiste I repeatedly stated here, and by now even a 5 year old child with severe cognitive deficiency should understand, honestly, that I'm NOT against learning by frequency and I always try to do it myself. What I am saying instead is that 6,000 words will NOT get you even NEAR to understand most shows, even simple ones, and entertain ANY type of meaningful conversation with ANY educated adult. And that's a FACT. And about frequency order, although, as I'm saying here by the zillionth time, I'm IN FAVOR or using frequency, many successful learners that are proficient in DOZENS of languages, like Steve Kaufman don't care AT ALL about that. And his argument is true too. He says you shouldn't mind the most common words because they will appear ANYWAY. So although I DO use frequency lists and stuff I tend to right away ignore the first 500 or so, because I'll learn them anyway reading right in the first months in a language.
@eliiste In other words I'm NOT saying you shouldn't study the most common 6,000 words. What I'm saying is that you FIRST ANKI deck in a language should have instead AT LEAST the 10,000 most common words or above. So you can START to play with the language. To really understand it you'll need the first 30,000 to 60,000 most common words. It's the NUMBERS I'm talking abut. Not they being in A frequency order. That's good. And note I said "A' frequency order not "the" frequency order. Because there's that too. Depending on your database the order will vary WILDELY! The MIGAKU deck, for instance, is a frequency order based on NETFLIX subs. That will have, of course, a lot of overlap, but also a lot o differences (specially in the ranking the word appear, not so much the presence or not of a word) if they were to use shows like the Abema Prime (which in my opinion would be way better) or light novels or REAL BOOKS, no fiction real books, or the news, or whatever. If you run the same program as they do you'll see words that appear right on top on one list moved WAY DOWN on the list in others, sometimes by hundreds or even THOUSANDS of positions in the first 10,000 words.
Unnecessarily long and repetitious IMHO. Sorry to make a negative comment but it's feedback nevertheless
How to make less long? Please help me
The repetition in the video is an intentional style to help understand the concepts to those new to it.
Well, just reduce the account of repetition of the 80/20 and word counts. I heard the 80%, 85%, 90% etc too many times. Please don't get me wrong, it was an informative video but I suspect the drop off rate is high in your stats. Thank you.
@@iosmusicman Thanks for the feedback, we will certainly take a look at the retention of the video and see what can be learned. 👍
@@MigakuOfficial I don't think the video is too long. It's perfect. It explained the importance of focusing on the most frequent words. All explanations were tied to the process of how Migaku recommends us to study certain words. It's almost a tutorial which I like...
It was a a well-done video.
If somebody dropped out just because they heard 80%,85 or 90% "too many times", they certainly don't have the grit necessary to learn a language.Your thoughts and statements exposed you @iosmusicman
You can't understand anything in a conversation with 80% comprehension.
Bruh did you even watch the video? It’s 80 percent of the words used in Japanese Netflix. which movies and tv shows obviously use different vocab then real life, but this is a very good list to understand tv shows and movies in immersion
Yes, also the key of learning by frequency is that first learning the top 80% of words starts to unlock a lot of content by making lots of sentences only have a single unknown word in them. These sentences then can be usually be understood when looking up the unknown word. Then you can focus on common words first again and continue to increase your vocabulary unlocking even more sentences. So yes, this video is about the approach of learning new vocabulary and is not emphasizing how much you will understand at 80% (or 85% or 90% for that matter).
@@orlando5366 This is what 80% comprehension looks like:
“Bingle for help!” you shout. “This loopity is dying!” You put your fingers on her neck. Nothing. Her flid is not weafling. You take out your joople and bingle 119, the emergency number in Japan. There’s no answer! Then you muchy that you have a new befourn assengle. It’s from your gutring, Evie. She hunwres at Tokyo University. You play the assengle. “…if you get this…” Evie says. “…I can’t vickarn now… the important passit is…” Suddenly, she looks around, dingle. “Oh no, they’re here! Cripett… the frib! Wasple them ON THE FRIB!…” BEEP! the assengle parantles. Then you gratoon something behind you…
@@spiritsplice yep you didn’t watch the video lol
@@MigakuOfficial I don't see any links to the japanese course you mentioned. All I see is a word list that i would have to add manually.