A couple months ago I was reading in this awesome forum and astridtops mentioned a "quick and dirty" method of studying RTK (in this thread #6) . Others have mentioned this method as well and from what I understand, the method involves letting the failed box pile up while one plows through the book as quickly as possible.
So first question is, can anyone describe in more detail this method? Or does anyone know the original 'quick and dirty' thread? (My searching skills are rot. Actually I can't find the milk in the fridge unless my wife is pointing at it.)
And secondly, what method do you use? What has been effective for you?
I will start with mine if anyone is curious. This is the order I do things in a day:
1. I clear out my failed kanji box (usually less than 10 cards).
2. Blue box (I never do the blue box on the same day I add them).
3. I review all expired cards.
4. I go to my new failed kanji and review the stories for the ones I failed.
5. I learn new kanji and add them to the blue box.
I like this approach because it feels systematic. The problem is I run out of time before I am learning new kanji. Doing #1-4 takes usually less than an hour. And I learn new kanji at about 10 per hour. (I have seen some say that 10 per hour is too slow. But I believe Mr. Fabrice himself (aka 'My Hero') seems to be in this camp as well). I am only able to study for less than 1-2 hours a day. I suppose I am getting frustrated with how long it is taking. At 900 kanji it is like I am having a bit of a Heisig mid-life crisis. Please help before I get sucked in by the dark side (no offense Miss Tham). Thanks in advance for the comments.

[ファブリス]

Hi laner36,

first off, some amount of frustration and burn out is perfectly normal and at the 900 kanji mark you're well on your way to completing this book as long as you don't take too long breaks from the study.

Your approach is fine, for the last quarter of the book I used the website for reviews (I started the book before creating the website), and I did the same, except that I would always study new kanji first. If some of the new kanji would re-use previously learnt components, it would serve as a kind of extra review of the blue stack ahead of time, and it wouldn't hurt given that the blue stack has only been learnt the day before.

Try switching your priority to new kanji first, and then failed kanji, and THEN review expired cards. Furthermore, give priority to expired cards in the first two stacks. Those are critical. Then you might want to "race" to 15 kanji a day during the week, and sometimes catch up on the failed cards and expired cards the w-e while taking a break from new kanji (forgotten ones don't take as much effort to rework on, as you are already acquainted with the writing and primitives).

Another reason for my suggestion, is that if your time is limited, it's better to keep your mind fresh for the new kanji, review is important but less demanding mentally wise.

Summary : prioritize new kanji > expired cards in stack 2-3 > other expired cards > failed cards.

You may swap the last two, and you could even wait for the weekend to do them.

EDIT: I think member Ziggr had posted a great schedule for this.

Which reminds me why being able to select a particular stack for review IS actually helpful, even if in a perfect world, it shouldn't be required.

Quote:So first question is, can anyone describe in more detail this method?

I kind of used the same method for the first half of the book. You'll have to trust your memory for this. Why not try it? Do kanji in bursts of 100, 200 or more, and review it all only at the end. I always did the 1st overnight review though.

I only _select_ a story during a study stage. The selected story can be as simple as a single phrase, without any associated imagery or details. Usually it's a borrowed story, which I might not like very much, but still I accept it as is. It makes studying quick.

Stories selected this way work without further elaboration in most cases, thus this is a good default strategy. I add details to the stories or elaborately imagine them during reviews if needed, for example if components get forgotten or keyword gets associated with story for another kanji, in which case details need to specifically differentiate between these keywords. This process is iterative, since story is already selected and only accumulates additional details.

As a result, about 50% of kanji go straight to last stack on the first try (I might be way off with numbers here...), 40% circle couple of times and receive minor additions to story imagery, 5% can circle several times in lower stacks until they irritate me enough to get a major rework of a story.

Cumulatively, it's a win, it works as a meta-Leitner system. In vanilla Leitner system, you _review_ only stories you have trouble with. In 'quick and dirty' approach, you also _study_ only stories you have trouble reviewing.

I'm attempting a "quick and dirty" method right now. The plan is to learn 100 kanji a day, for three weeks, without fail. I just finished my first week today and uploaded my 700th kanji moments ago.

So far there is no backlog of kanji, but should a stack of expired cards grow out of control, I don't plan to worry. I'll just review as much as possible in a day while retaining my 100-new-kanji-a-day pace. Seriously, after finishing my speed run [1.5 to 2.5 hours a day learning new kanji] I'm going to have plenty of time to dedicate to reviewing and clearing out expired stacks.

I know I'll be retaining far less kanji than normal at this pace, but I'll be done so quickly that I can put all my efforts into reviewing and mastering the kanji that didn't stick during my speed run. As others have said, failing isn't a bad thing. What I'm seeing during my reviews is that I frequently remember keywords by failing them. It's hard to describe, but getting it wrong and then seeing the answer cements it in my head.

It's because of all this that I believe people are too quick to jump on the Leitner train. Despite its inherent time-saving capabilities, all flashcard systems have serious disadvantages. When reviewing 2,042 (!!) cards, so much time is spent reviewing that progress slows to a crawl. It just feels like there is some step missing in the time optimization process right now. I can't figure it out myself, but maybe the "quick and dirty" method will help.

Anyway, I'll update with my progress as I go.

EDIT: For those wondering, I'm remembering about 65% of new kanji the following morning.

Talka, 65% of 100 kanji remembered the following day when you are learning 100 kanji a day sounds like a very good retention rate. Although from reading wzafran's report of his 100-kanji-a-day project you may find this rate takes a toll on you over the next two weeks. (You may want to check out his thread.)

Still I think you may be right about doing a "speed-run" through the initial learning phase so that you have more time to concentrate on review afterward.

Let us know how it goes. Especially your retention rates after you finish RTK1 and your kanji start progressing though the boxes. It will be great to have data on this. Perhaps wzafran can give us an update on his retention as well.

Good luck to you!

Just an update on my 100-a-day "quick and dirty" speed run. I'm still on track, and plan to upload my 1500th kanji in 15 days tonight.

I might be hitting a wall, however. Normally I retain about 50 to 75 of the kanji I learned the day before. Today, however, I remembered just 35. Man, that hurts. You spend 3 hours studying one night, and only 1 hour of it sticks. Geez. Additionally, I've got 300 kanji in the failed pile at the moment (I'm been trying to keep it under 200, tops). Reviewing those failed kanji and learning the newer complex kanji at the same time is tough, to say the least.

To those thinking about doing a quick and dirty method: it'd be worth it, for your sake, to choose a 50 a day pace. Everybody's maximum speed is different, for sure, but 100 is just so bloody many to do everyday. 50 kanji in a day is just so much more reasonable and so, so very possible. I have to keep up my pace for time reasons, but if I could do it again I'd have gone at a 50 a day pace (and started 3 weeks earlier to get it in). But who knows, maybe when I'm done I'll be glad I got through the hard part in just over 20 days. Reviewing failed kanji is so relaxing compared to learning new kanji en masse. It honest-to-God feels like I'm taking a relaxing break when I do the reviews.

For anybody out there who's done it at this pace or is doing it at this pace now: I understand your pain, dude.

If you want to learn the meanings to the joyo kanji as fast as possible, you could try Slime Forest Adventure. It really worked well for me... I learned about 1000 kanji in about 10 days, with a good retention rate. I think it took 1-3 hours a day...
Anyway, there's a lot of problems with using SFA; it usually uses different keywords than RTK, many kanji have the same keyword, the order is different from RTK(first 1000 kanji are grouped with look-a-likes, so that helps, but the rest is random last time I checked...)

Hah, when I used it, I just walked back and forth in the forest :B I just find the repetition made things stick easily.
But checking the Slime Forest Adventure site now, it looks the the owner changed the system. The free version only has a few hundred kanji. Oh well.

Chadokoro_K Wrote:Still I think you may be right about doing a "speed-run" through the initial learning phase so that you have more time to concentrate on review afterward.

Getting through the kanji fast feels great, and it seems like you are conquering a significant portion of Japanese in a very short period of time. However, the only thing I would warn you about is that once you get towards the end, it is very easy to say "I'm done" and move on to reading real stuff in Japanese and finally using the knowledge you have gained. I was so impatient (because dental school was starting in about a week) to finish RTK and actually start using my kanji knowledge, that I ended up speed racing to 1800 kanji, then moving on. That left me without knowing over 200 commonly used kanji, and also not having a solid grasp on kanjis 1600-1800, which I hadn't reviewed very much. Now, looking back one month later, I want to go back and finish RTK, but the 1000 or so expired kanji seems very daunting.

If you can keep up with your reviews, however, faster is definitely better!

I started out making the entire list of joyo kanji from scratch. That took forever, then I imported to the old Memory Lifter Program. Since then anki has come about and I love it. I've never used this site to study (Nothing personal Fabrice, just don't have on-line access all the time. Love the forum, though!) so I'm unfamiliar with the system, but from the get go, I worked a chapter at a time. I always kickstart anki and knock out all my reviews first. Once caught up, I start my new stories for the next chapter. Once I've typed them up in Word, I might glance over the chapter again or not, then continue anki until I've passed the new chapter into the system. At this point, over 1800 now, creating the stories takes up most of my time. As others have said, the reviews are painless in comparison. Some chapters were pure agony because they were so long, but just keep plugging along. What else can you do with 2042 kanji besides give up?

If I had more restrictions on my time, perhaps I'd just sit and write tons of stories. And if we're tweaking/cheating the system anyway, why not then enter the kanji while reading the stories for each one. If you've gone through the trouble of looking at each kanji and creating a story(or finding one), I consider that exposure and study. Then flipping the cards counts as double exposure, so to speak. At the time I miss a kanji, I look it up either in the book or in my story file or both and also with the anki editor look-up function. This is more exposure (over exposure? ). If kanji comes back around in the same session, I always give myself a lower grade to be fair.

So, as long as your consciously thinking about the kanji, its parts and the stories and reviewing, I don't see any problems with high numbers. Heck, do 500 a day and you'll be done by the weekend! Good Luck!

This thread has kind of gone in another direction, but I wanted to say a huge thank you to those who helped me last month.
Fabrice, I don`t know why I didn`t think of learning the kanji before reviewing the kanji. It is so simple but so genius. I did as you suggested, and put my priority into learning new kanji first on M-F and on the weekend, catching up on failed and reviews from the 3rd and 4th stacks. It has worked great!
Chadokoro--I started _selecting_ stories as you suggested and it has greatly sped up the time to learn new kanji. I was a bit shocked at first by my horrible failure rate. But I think you are right in that it will be cumulatively better.
Talka--thanks for your inspiration. I figured if you could do a 100 a day, I should be able to manage 20 a day.
My progress since then: I have kept to 100 new kanji a week. So by friday I will be over 1500! My Christmas present this year will be finishing RTK 1.
And Finally, thanks again to Fabrice for setting up this site and those who write in this forum, without which, I would have given up ages ago.

[ファブリス]

Congrats laner36, keep it going. Don't overburn yourself though, 20 a day, or 100 a week is a very good pace.