Hi all. I’m very new to Japanese, only recently having finished memorizing kana, now working my way through an Anki core deck. I love Anki, but I’m a bit lost as to how I should best use my study time. The consensus online I seem to get is that I should be ditching the core deck as soon as I can in favor of a self-made “mining” deck, made from words I don’t recognize when immersing. My question is, when can I know that I’ve hit this point? How soon should I be immersing and mining?
I’ve already been attempting to read Japanese material, but I’ve had to look up nearly every single word. Does this mean I should be mining too? Or should I just postpone reading until I’m more proficient with the core deck? Thanks!
5 comments
I think mining works best when you don’t need to look up every single thing in a given sentence, sentence by sentence. I never did any of the Core decks so it took a very long time to meet that condition. Five years of just absorbing what I hear so often without any form of review except or hearing those words again and again in context, to be precise. I started mining three years ago, and it definitely sped things up.
So in your case, you may wanna keep up a Core deck for a while, though you might wanna go for, say, 2k instead of 10k. If you choose native materials appropriate to your level, the 2K would suffice as a start, at which point you can start mining. You’ll know you should start mining when you mostly get the sentences except for maybe one or two key words.
Opinions on this vary greatly, from some people starting mining after their first 500 words to others learning 6,000 words first. Do whatever works for you, but from what I have seen in previous discussions, the average person learns ~2,000 words from fixed lists before starting to mine.
That is going to vary on what materials you are studying and your own style. If you read a lot of native material, you probably wouldn’t even need a mining deck (I didn’t).
To be honest I’m not a fan of mining approach, and I think that people are trying to invent the wheel. I mean, what’s the point to create something that many people have already done and polished before? To make it worse, sometimes people literally add all unknown words and end up with a 10k deck in amount of weeks.
The way I do it, I don’t add any common word, because I will learn it naturally over time. If I want to add some card into SRS, then it’s something that both rare and interesting at the same time, so I know that I won’t see it very soon in a natural way, but I still want to learn it. Similarly at beginning stages you can expect to learn around 15 words in a hour naturally. So if you aim to learn something like 20-30 words and you use content for several hours daily, you don’t even need SRS so much from the very beginning.
There are good and bad sides with such approach. Good point is that you don’t need to spend any efforts on creating cards or even SRS at all. Bad side is that if you learn words in a natural order then the content you use affects what you learn a lot. A simple example, if you read fantasy books, then elves and orcs appear hundreds of times while words like office either 1-2 or none at all. Because we are talking about literally thousands of words, it’s actually possible to learn for 1-2 years, get extremely proficient at reading fantasy books, but still don’t know even very basic words. But it’s a problem of content learning generally comparing to learning words based on frequency and so on. I suppose if you create cards, you might be able to sort it by frequency and learn in a frequency order, so even if some word appears only once, but it’s very common in other settings, you still will learn it first.
Mining is exhausting when you don’t have much of a vocabulary. Jumping straight into mining now is going to a lot more difficult and tedious than if you wait until you have “learned” 1000-2000 common words and then go for it.
You can do it either way, but not knowing words or grammar structures is going to tire you out.