Hi RtKers, I've downloaded "Core 2k/6k Optimized Japanese Vocabulary" and I've seen the default layout is like:

---

FRONT

that, that one

( )はとってもいい話[はなし]だ。

That's a really nice story.

BACK:

それ

それはとってもいい話[はなし]だ。

---

1) Why in the default layout furigana is shown between [ ] and not in furigana form? Is there a specific reason or I can safely change this from {{Sentence-Clozed}} to {{furigana:Sentence-Clozed}}? (sorry for the excessive prudence here )

2) How it is supposed to work? When I see the front side, I have to write the corresponding kanji compound (if there is one) and write or pronounce the reading for the word?

I know those are dumb questions but before I start with this deck I do preffer to guarante myself I'm doing it the right way

BTW great deck, good job to the author (I think it is Nukemarine, sorry if I remember wrong)!

EDIT:

I must add I'm more interested in recognition that in production, plus I fear it's not healty to translate from english to japanese, so I was thinking to change the card's format and put the japanese word on front in kanji, then in the answer I'll put the reading in kana and the meaning in english.

My fear is that in this way I won't memorize the kanji compound as well as in the defaul format, for the same reason in RtK is better to go from keyword to kanji. :/

Maybe on front I can put the reading in kana, so I must anyway recall the kanji, but without starting from english. But then will my retention of the reading be affected? When in the real world I'll encounter the kanji compound, I'll find difficult to recall the reading?

For the studying of it, I'd just recommend redoing the card format so that it's recognition instead of production.

As for the furigana, if I remember correctly, putting it in brackets used to work [ひらがな], but doesn't now, so to fix it, you'd first have to find the new way of doing it (which I don't know, 'cause I never bothered to change it) and change the card format so that it does as it's supposed to.
You'll notice that the furigana for the sentences will work, so I guess you should start with trying that format.

Or you could wait for someone who knows a bit more about Anki to tell you what to do. I only switched things around to make recognition cards, so I don't really know all the formatting options.

EDIT: Looking at your post again, I realized that the sentences weren't showing furigana either... The rest of the advice stands though, find the way it works now and edit the card layout so it'll work.

Personally, I don't like production cards; they take a lot longer to study, which, at your level, makes them nearly useless. You might try some form of production cards later, but for Core 2k (and even 6k, in my opinion), they really aren't worth the effort. It's up to you though, you might get better results than I did.

Thank you sholum for the reply!

sholum Wrote:As for the furigana, if I remember correctly, putting it in brackets used to work [ひらがな], but doesn't now, so to fix it, you'd first have to find the new way of doing it (which I don't know, 'cause I never bothered to change it) and change the card format so that it does as it's supposed to.

Ahhh ok! I had been wondering if there was a reason for this. So I changed it to furigana

You have to change this:

{{sentence}}

to this:

{{furiganaentences}}

and the content between squared brakets will be shown as furigana.

sholum Wrote:Personally, I don't like production cards; they take a lot longer to study, which, at your level, makes them nearly useless.

I was thinking the same thing, also because I am using Anki especially when I am in bed or in general in other situations where it is difficult for me to write down on paper (I know ankidroid has the virtual white-board but imho it's not the same thing as paper and pencil), so production might be not ideal for me.

The only concern I have is that recognition only may not imprint the kanji compounds in memory in the same intensity as production might do. I don't know... for RtK I've tried kanji -> keyword and I've seen for myself it's not as effective as keyword -> kanji... with keyword -> kanji is like the kanji has been imptinted in your mind while in kanji -> keyword it is more volatile and superficial... but at the same time I must admit the second takes 1/5 of the time.

For example I've inserted a and the respective sentence from a song I like, in recognition only:

沈黙

深い沈黙 どれくらい続いたろう
TVの音がやけに大きく聞こえる
どんな言葉をずっと待っているの?
小さな部屋が 広く思えた

now, considering I am listening to this song like 10 times a day xD I memorized both reading and meaning, but I feel like those word is prone to vanish or to become cloudy in my mind... while I'm pretty sure if I do production with this, it will be branded indissolubly in my mind... so I don't know what to do :/ Sorry if I bother you with my stupid indecisions xD

I'm using this same deck although I exported it to another SRS program (Flashcards Deluxe). I tried it with production and while it's a great way to exercise your Japanese I don't think production works so well in an SRS. After a week you might not remember just which word it is that fits etc. I had a cool idea in that I changed it to make it show the sentence minus the target word and the audio on side 1 so that I wrote the missing word using the correct kanji etc. This was a lot of fun and I think it's doable but also it's slow. You'll realize that you could be learning vocabulary faster if you just do it the standard reading the kanji way. I'm going to make myself learn to write Japanese somehow someday but not sure when.

I like that deck a lot though.

I've also found that using recognition cards is a bit more volatile, but I think that the increase in speed is worth it. Generally, I find that a word either sticks quickly or not at all (it usually sticks); however, if that word is actually useful, I'll run into it a lot while reading and reinforce it that way (the only way I could remember 撃つ). That's why I'm okay with only using recognition cards in Anki.

If you want to work some production into your routine, you could always study the ones that seem the most volatile from a list; from what I've heard around here, you only need to actually write the word two or three times to get most of the benefit from it, which is another reason I think production cards are unnecessary.

Or, if you can find a copy (or 'find' a copy), you could try 正しい漢字かきとりくん (either version is fine; I prefer the drill system in the first one, but the second one has better features overall). It's pretty entertaining and serves its purpose pretty well.

For #1, I thought I had it set up so that no furigana shows up on the question. It's not essential to remove it, just something for those that want to test their ability to read the entire sentence including the clozed portion.

For #2, yes, the idea is based on the Clozed sentence, English translation, English keyword, image and whatever other hints you personally added that you can from memory write the correct Kanji form of the word and know its pronunciation.

Personally, though it would take more work, it'd be worth it to add images that takes away the need for the sentences and the English words. That becomes difficult for abstract concepts but is still doable. I did this for a Spanish deck of DuoLingo and found it very useful. Basically, can you figure out the word from the Picture, if not then look at the clozed sentence, if not look at the English word/sentence. You're not locked into just using the image as the clue, but the image is the primary clue. This concept to target word has lots of supporters.

This was just off personal experience from a few years ago that passive reviewing where I'm just recognizing the Japanese word or the Kana version of it was not that good at all. I had a hard time recalling it when I needed these words for conversation or writing.

PS: My Japanese level is now horrible given that I haven't studied in any serious fashion for over 3 years now so take my advice with a grain of salt. There was a legit reason behind the original set-up so I stand by that.

If you intend to learn Japanese by reading (and are only using this deck to make reading easier), then I think recognition might be enough. But you should take advantage of the "optimized by sentence" index in the deck, if you plan on doing it that way. Don't do it in the current order, which is optimized by vocab. But, in my experience, for anything else (even listening comprehension), doing the clozed deleted production version helps a lot.

Another option is to create a second set of cards, and do both recognition and production, simultaneously (one in optimized by sentence, the other optimized by vocab order, to help break up the monotony among other reasons).

I would suggest creating two decks (create the second set of cards in the original deck, copy them into a new deck, and then re-order the new deck to "optimized by sentence" - I'd be happy to describe how in detail if anyone needs it), and doing them in parallel. It will make it easier to choose which type of cards you want to add, at any given point. (you might want to stay ahead with the comprehension deck, and use the production deck only as a means to reinforce some of the difficult words that don't seem to stick with the other method).

Eventually you'll likely drop one, and stick with the other, but, at the beginning, doing both might make things go smoother (fewer failed cards), and help you really learn the basics (assuming you're a relative beginner). It would also give you an idea of which method you prefer, help you make an informed choice.

And it's not like suspending cards that are too easy isn't an option. So, with that in mind, doing the deck twice might not take up as much extra time as one might expect. There's absolutely no reason to "produce" English loan words for instance - even if you wanna practice your katakana, recognition is better for that.

Thank you all for your replies! I think for now I'll make three cards for every word/sentence: kana to kanji, kanji to kana, cloze deletion (with images and english as Nukemarine suggested) so I can see what works better for me and eventually I'll take advantage of all three for some words and suspend the other

Kana to kanji? You mean write down the sentences every time?

Stansfield123 Wrote:Kana to kanji? You mean write down the sentences every time?

Not every time, only the first two or three time, then I can visualize it in my mind... I know it's not like writing it down everytime, but I'm still activating my mind to recall the kanji and I think this is good for memorization...

Or when there is a kanji i cannot recal, then I will treat it as it's new and write it the next time I see the same kana...

I don't know if this works in the long time, maybe I can do it only for some compounds I find difficult to recognize when I see them...

For now I do only recognition, then if I encounter a kompund I am unable to memorize, I'll unsuspend kana to kanji and cloze deletion for that compound

Be wary of kana to kanji as that also is passive. I did it in my studies as an offshoot of AJATT, but it is a problem since I had less ability in knowing the readings of Kanji. Reason being, the image and sample sentence gave me almost enough info to know what the Kanji should be, however the sentence also had the kana pronunciation there on the question side. It made it almost too easy. Even removing the images didn't help because the kana sentence gave enough clues. Really, I was just training my ability of how to write kanji which I was already doing with my RTK deck.

When I did cloze delete, that changed things. Now I needed to know how to produce the kanji and the pronunciation. It was much more difficult, but the benefit was much more tangible as well.

Nukemarine Wrote:Be wary of kana to kanji as that also is passive. I did it in my studies as an offshoot of AJATT, but it is a problem since I had less ability in knowing the readings of Kanji. Reason being, the image and sample sentence gave me almost enough info to know what the Kanji should be, however the sentence also had the kana pronunciation there on the question side. It made it almost too easy. Even removing the images didn't help because the kana sentence gave enough clues. Really, I was just training my ability of how to write kanji which I was already doing with my RTK deck.

When I did cloze delete, that changed things. Now I needed to know how to produce the kanji and the pronunciation. It was much more difficult, but the benefit was much more tangible as well.

You make me reflect... In fact I was suspecting this and you give me the confirmation... like in the sentence I put before:

FRONT:

word = 沈黙

sentence = 深い沈黙 どれくらい続いたろう
TVの音がやけに大きく聞こえる
どんな言葉をずっと待っているの?
小さな部屋が 広く思えた

BACK:

meaning = silence

But I've seen the sentence is making this too easy... In truth I'm inferring the meaning from the context of the lyrics... I suspect if I see the same word in another context I won't be capable to read/understand it...

So in the end I think you're right and I'm going to do what you have said:

FRONT:

meaning = silence

sentence = 深い[...] どれくらい続いたろう
TVの音がやけに大きく聞こえる
どんな言葉をずっと待っているの?
小さな部屋が 広く思えた

BACK:

word = 沈黙

Thank you Nukemarine, I think your advice will make me gain valuable time in the long run

But I have other two question for you xD

3) Do you think this format is good only for vocabulary or also for grammar? Or for grammar it is better to stay with the classic sentence -> meaning? So I read the sentence in japanese and I try to understand the meaning; then I look at the answer and I check if I've understood the sentence and guessed well the meaning.

For example for now I'm studying Genki I to learn basic grammar (conjugations, particles, sentence patterns etc..)? How would you put the sentences from this book in Anki?

4) If a sentence has more compounds I want to learn, I've to put the same sentence several time? Like:

meaning = deep, profound

sentence = [...]沈黙 どれくらい続いたろう
TVの音がやけに大きく聞こえる
どんな言葉をずっと待っているの?
小さな部屋が 広く思えた

---

meaning = silence

sentence = 深い[...] どれくらい続いたろう
TVの音がやけに大きく聞こえる
どんな言葉をずっと待っているの?
小さな部屋が 広く思えた

(Obviously those questions are for self-made decks, not for Core2000)

EDIT: to rephrase my first question, do you think I can take advantage ofcloze deletion with the sentences I'm mining from Genki? And if yes, what do I have to cloze delete? Thank you in advance!

Sorry to flood you with so many questions xD

5) What would you put as a hint for the clozed sentence?

This:

hint = eat;

教室で[食べる]の。

Or this:

hint = I eat in the classroom.

教室で[食べる]の。

?

Stansfield123 Wrote:...There's absolutely no reason to "produce" English loan words for instance - even if you wanna practice your katakana, recognition is better for that.

You might be right some of the time with the words that can easily be converted into the Japanese syllabary. But thanks to the Japanese habit of cramming every word--no matter its phonetic composition--into a (very) limited set of sounds, a lot of words you do need to practice. At least, if you ever want a native speaker to understand you. If you never intend to speak Japanese, I guess it isn't a problem. Not sure why somebody would go that route, though.

A recent example off the top of my head:

ウィルス

Oh how I love practicing how to horribly mispronounce my own language.

afterglowefx Wrote:A recent example off the top of my head:

ウィルス

Oh how I love practicing how to horribly mispronounce my own language.

I don't know what your native language is, but just in case, ウイルス isn't taken from English (but I can't remember what language it's from).

That brings up a good point however, many loan words aren't from English and any resemblance is simply due to the fact that English shares that word with the actual source language (which might be the source for the English word as well). As such, unless a katakana loan word is obvious to me, I study it. Due to the phonetic nature, I usually remember it quickly (I can usually still hit 'Easy', but I make myself look at it more), but either way, it's good to go over them.

sholum Wrote:

afterglowefx Wrote:A recent example off the top of my head:

ウィルス

Oh how I love practicing how to horribly mispronounce my own language.

I don't know what your native language is, but just in case, ウイルス isn't taken from English (but I can't remember what language it's from).

That brings up a good point however, many loan words aren't from English and any resemblance is simply due to the fact that English shares that word with the actual source language (which might be the source for the English word as well). As such, unless a katakana loan word is obvious to me, I study it. Due to the phonetic nature, I usually remember it quickly (I can usually still hit 'Easy', but I make myself look at it more), but either way, it's good to go over them.

Ah, a bit like エネルギー then, which both Japanese and English took from German.

Regardless, even with words that pretty closely match English, trivial pronunciation errors lead to massive confusion. Just as an example, I bought my car here at auction. I get asked about it a lot, because it's a fairly rare car here in Japan. For months I was telling people I bought it at アクシュン, instead of the actual Japanese loanword オークション (coincidentally nobody seems to use 競売).

I don't think a single person understood what the hell I was talking about the first three or four times I had to repeat it. Then, finally, their face would light up, and they'd get what I was saying. This happened over a dozen times. So yeah, I definitely keep my loan-word production cards around, because trivial things like, you know, being understood matter to me. I still say ドラ instead of ドル all the time -.-

afterglowefx Wrote:Ah, a bit like エネルギー then, which both Japanese and English took from German.

For the record English didn't take engergy from German (etymonline.com):

energy (n.)
1590s, "force of expression," from Middle French énergie (16c.), from Late Latin energia, from Greek energeia "activity, operation," from energos "active, working," from en "at" (see en- (2)) + ergon "work, that which is wrought; business; action" (see urge (v.)).

Virus is a Latin word. But I don't know where Japanese got it from. Russian maybe (thinking of ウォッカ obviously)? The -i in virus is different from the -i in vision, maybe the katakanisation in　ウィ　instead of ビ　just comes from here?