...For you?
I'm just curious about how people define the term "finishing RTK".......
Is it when your retention rate is like above 90% for all cards? Or is it when you've finished selecting your stories for all 2042 kanji? Or is it when all of the cards are "mature" in Anki?
There might be a general convention about what finshing RTK means, which I skipped somewhere...could sy enlighten me?

I think once you've gone through all 2042 kanji or 3007, which every you want to do. and you've taken your time. like 3+ months. the rest the srs will do. RTK is a means of learning to associate meanings with each kanji/write kanji in general. Once you've done kanji and reached a considerably level of japanese. You won't really even bother thinking about keywords, unless you plan on learning more kanji then 2042 or 3007. Which I am for personally.

It doesn't mean anything special to me, I guess. It took me 10 months to finish (finished in Dec. 09). I've said it before on this forum, but if someone would've told me at the start of it all that I would actually have stuck it out for almost a year and finished, I wouldn't have believed them. Before RTK, I also came from zero Japanese background (no classes, didn't even know what kana or kanji were). The only Japanese I had been exposed to in my life came from music I listened to and other media.

I don't care for retention rates all that much to be honest, but I'm biased because I hated most of RTK after the halfway point and still loathe my reviews everyday. For the record, I average between 80-90% retention on most days. Some days as low as 75% and others near 95%. For the people who can do 90%+ everyday, my hat's off to them. Maybe I'm stupid?

Most of the kanji I fail on my RTK reviews are due to keywords being nearly identical anyway, and because I just never see them in the wild yet. The few kanji I have seen countless times already just from Tae Kim and ~100 sentences into KO are cemented into my mind pretty easily when I review them on here. I actually almost wish I didn't see the English keyword for some of them because it's easier for me to visualize the single kanji by picking it out of a compound/word I've already studied.

As for going beyond the 2042/3007 kanji, I'm for it. Personally, I don't even use keywords for the non-RTK kanji that I actually bother to add into my SRS, and so far, I remember them easier than some of the RTK kanji

Raschaverak Wrote:...For you?
I'm just curious about how people define the term "finishing RTK".......
Is it when your retention rate is like above 90% for all cards? Or is it when you've finished selecting your stories for all 2042 kanji? Or is it when all of the cards are "mature" in Anki?
There might be a general convention about what finshing RTK means, which I skipped somewhere...could sy enlighten me?

Each point you listed I think I used as "milestones" to post in the "I've just finished" thread.

1. Get all stories into rotation ie reviewed at least once (MUCH more difficult back in the days when each kanji may have 1 or 2 stories max).
2. Reduced to missed decks down to zero on a daily basis (I had the bad idea to keep adding new cards without reducing my missed deck)
3. Gotten at least 90% of RTK into mature stack (Jan. 2008 baby!!!)

Personally, once you have 90% as mature, I think you count as 'finished' but that does not mean finished as in stop reviewing. That's finished as in RTK is not your main concern anymore. Time to move onto finishing the next milestone whatever that may be.

Why 90%? Well, I work for the US government and 90% fits into a saying we have: "Good enough for government work." Plus, perfection can be the enemy of progress.

For me it was when I added my last stories. Of course, RTK doesn't end there at all, you got reviews and whatnot. But now I so miss the good the times when I was making stories. It makes me want to do RTK3, just for the fun of it! And perhaps I will, if I just get the time..

paasan Wrote:For me it was when I added my last stories. Of course, RTK doesn't end there at all, you got reviews and whatnot. But now I so miss the good the times when I was making stories. It makes me want to do RTK3, just for the fun of it! And perhaps I will, if I just get the time..

Yeah, well I've made my own stories for a lot of kanji, and I regret it deeply, it was a waste of time. If I were to do RTK again from scratch, I would just copy the most starred stories from the site (maybe without even reading them), and start srs -ing immediately....that would be the fastest way in my opinion. If you can make up your good story in a very short amount of time (like under a minute)...well that's good. But putting too much time in making stories is just not worth it

Raschaverak Wrote:

paasan Wrote:For me it was when I added my last stories. Of course, RTK doesn't end there at all, you got reviews and whatnot. But now I so miss the good the times when I was making stories. It makes me want to do RTK3, just for the fun of it! And perhaps I will, if I just get the time..

Yeah, well I've made my own stories for a lot of kanji, and I regret it deeply, it was a waste of time. If I were to do RTK again from scratch, I would just copy the most starred stories from the site (maybe without even reading them), and start srs -ing immediately....that would be the fastest way in my opinion. If you can make up your good story in a very short amount of time (like under a minute)...well that's good. But putting too much time in making stories is just not worth it

I find that view both impractical and a bit selfish. If everyone thought like this, the site would be pretty bare, wouldn't it? If the method worked for some stripped down universal set of stories, wouldn't be much point for this site. You want to find a good mix of finding what works for you from shared stories, not all of which will be the 'most starred', and coming up with your own, and preferably sharing them. Some stories will take longer than others, some will be very minimal, some will be long, but what's important is sound logic and effortful use of imagination according to what resonates with you as an individual.

Nukemarine Wrote:perfection can be the enemy of progress.

This is so true....trying to be perfect, or trying to make sg perfect is usually not worth it..
for every reached new percentile you have to give more and more energy / time whatever... which is a waste. The clever thing would be to find the optimum what you should aim to achieve with your resources available, and then move on...
a major principle I've overlooked when doing RTK.... just thinking about it makes me sick

nest0r Wrote:I find that view both impractical and a bit selfish. If everyone thought like this, the site would be pretty bare, wouldn't it? If the method worked for some stripped down universal set of stories, wouldn't be much point for this site. You want to find a good mix of finding what works for you from shared stories, not all of which will be the 'most starred', and coming up with your own, and preferably sharing them. Some stories will take longer than others, some will be very minimal, some will be long, but what's important is sound logic and effortful use of imagination according to what resonates with you as an individual.

Point taken. I did share my stories by the way...although judging by the stars given on them, they weren't much of a success among the members
The point is, that the most starred stories should work for most of the people in most cases (kanji). That's it. What you're saying is, going in the way of being "perfect" and coming up with the best / better stories for kanji (if you need to). If you are not using an srs this might be a valid idea. But I think the srs takes care of the rest of the problem....
I never mentioned that making up stories is bad, or a bad idea. Nope. If you have the time, and enjoy it that is...but if your time is limited then the most efficient method would be what I've mentioned, I think. After all RTK is just a frameset for learning kanji, taking it too seriously will wound anybody up with wasting a lot of time on it...trust me I went this way and it doesn't feel good

I don't think 'most starred' means 'works most of the time for most people' at all. I also think the most efficient way of completing RTK is not half-assing it and relying on the SRS + rote memorization, but making sure you've got a good story to form the 'dissolving stitches' for the kanji-learning process. That's what makes the method so effective, internalizing the kanji from the bottom-up, making them your own, piece by piece, right up front, and doing it properly so that 'relearning' is just a matter of occasional, simple corrective maintenance in context.

I spent 6 months doing RTK at a rate of 20-25 kanji a day, sometimes taking breaks. I took my sweet time and had fun making stories, sometimes using others' stories and then later switching to new ones... when I 'finished' RTK, it was a matter of adding the newest batch of cards, with mostly mature cards in my deck. I didn't enjoy using 'starred' stories just because they were starred or using stories that didn't really resonate with me, or trying to rush myself. By the time I finished, I didn't even need to keep on with RTK, instead I just did kanji in words and sentences, I made the orthography my own and didn't sweat kanji from then on.

If I could go back, all I would change is the sort of thing I mentioned here, and I probably wouldn't have taken as many breaks--in other words, combining other components of Japanese learning in a more condensed span of time: http://forum.koohii.com/showthread.php?p...6#pid81476

I think "finished with RTK" means you got a story for all the Kanji and if you are using a SRS you have studied the card at least one time. That's just me because obviously we have to keep reviewing the stories to make them stick.

About the perfection thing. I totally agree. I've gotten to the point where I don't really care if I have left over RTK reviews by the end of the day because honestly it's not that important (as long as I keep reviewing enough daily so that it never gets overwheliming) basically I find that actually studying Japanese is much much more enjoyable and fun and as long as I keep RTK minimally (like I said, to keep the reviews in check while still giving them some "freedom") then I feel I have been keeping true to the way the book should work. Don't let RTK get in the way of your studying. if you have reviews to do but you would rather study real Japanese do it and worry about the reviews later.

nest0r Wrote:I don't think 'most starred' means 'works most of the time for most people' at all. I also think the most efficient way of completing RTK is not half-assing it and relying on the SRS + rote memorization, but making sure you've got a good story to form the 'dissolving stitches' for the kanji-learning process. That's what makes the method so effective, internalizing the kanji from the bottom-up, making them your own, piece by piece, right up front, and doing it properly so that 'relearning' is just a matter of occasional, simple corrective maintenance in context.

I agree about the "most starred' means 'works most of the time for most people" thing.
I can't know that, if it's true or not. However if I extend it a bit: "works most of the time for most people using an SRS" then I think we've gotten closer to the truth...
Seriously, 6 months is good, however keep in ming that Heisig suggests himself, that if sy is doing it full time (let's say 8-9 hours) then the whole thing can be done under a month... 6 months doesn't mean anything at all, without saying how many hours a day you've spent doing RTK. But then again, I don't think Heisig took into account the cool koohii.com site, with a bunch of shared stories and an SRS. He meant the whole thing without those additional tools. I bet if you asked him again with that little piece of information added, he might say, a week

Although many of the "most-starred" stories are really memorable, you have to account for how long they've been on the site. MANY of the top starred stories were created and posted as far back as late 2006 into 2007. That's 4 years these stories have had time to accumulate votes. Not saying they aren't good stories by any means, and there are plenty of recent stories that are top starred, just pointing out that time is indeed a factor for how many votes these stories currently have.

Raschaverak Wrote:

nest0r Wrote:I don't think 'most starred' means 'works most of the time for most people' at all. I also think the most efficient way of completing RTK is not half-assing it and relying on the SRS + rote memorization, but making sure you've got a good story to form the 'dissolving stitches' for the kanji-learning process. That's what makes the method so effective, internalizing the kanji from the bottom-up, making them your own, piece by piece, right up front, and doing it properly so that 'relearning' is just a matter of occasional, simple corrective maintenance in context.

I agree about the "most starred' means 'works most of the time for most people" thing.
I can't know that, if it's true or not. However if I extend it a bit: "works most of the time for most people using an SRS" then I think we've gotten closer to the truth...
Seriously, 6 months is good, however keep in ming that Heisig suggests himself, that if sy is doing it full time (let's say 8-9 hours) then the whole thing can be done under a month... 6 months doesn't mean anything at all, without saying how many hours a day you've spent doing RTK. But then again, I don't think Heisig took into account the cool koohii.com site, with a bunch of shared stories and an SRS. He meant the whole thing without those additional tools. I bet if you asked him again with that little piece of information added, he might say, a week

No, having the SRS doesn't make 'most starred' = 'works most of the time for most people', not sure why you think that logic makes sense. ;p

We've already had plenty of conversations about how long it might or might not take to complete RTK and how we've improved the method since Heisig. That's an old topic. As I said above, the only thing I'd change would be to compress the amount of time and how I used that time. As I said, I did 20-25 new kanji a day. The time varied, depending on the kanji and stories, but as I progressed, it took less and less time. That's because I put in the effort and enjoyed myself and 'mastered' the method.

What I would never do is rush myself or half-ass my stories and pretend the SRS was a magic pill that replaced the need for having stories that worked for me, whether they were long or short or starred or not. I don't know anyone who has completed RTK and after moving on would say they just used the SRS and stories that didn't really work for them to complete it. I've seen literally one or two people make claims like that in contrast to myself and many others, but I have no way of verifying what they say--especially about their (real, not measured, which is why I can't know it) error rate and effort invested, and I would tell them they could've saved time and effort doing RTK properly and moving on.

As a bonus, much of what I learned, in terms of metacognition, doing RTK, I developed and applied to Japanese in other areas.

The fault in the logic is assuming that stories get starred because they work for people, when in fact people tend to star the stories when they first choose them, and not after they've learned the kanji. The SRS does nothing to change the fact of whether or not the story is a good one. The SRS does make it possible to learn anyway, whether the story is good or not, but the question is: Does using a "good" story and the time it takes to find/make one save or waste more time? Because in theory, "good" stories should make the review process in the SRS much easier and efficient. Therefore, just because the SRS makes it possible to learn kanji using mediocre stories, it also amplifies the benefits of a good story just as much. That is in theory, however.

In practice, I don't think the stories matter that much as long as you have a loose logical structure to tie the keyword to the primitives, but I'm in the minority on this. I've found that stories I invested a lot of time/effort in are not really any easier to remember than simple, one sentence stories I put together without a second thought.

That still doesn't support your argument as you made it, but it does agree with you in sentiment. Spending a lot of time on stories only slows you down.

@kendo - I think you're in the minority on it because it's not a very good method (to start with), but that's just me, and read on for clarification. ;p Perhaps you miswrote, but I think what's important isn't tying keyword to primitives so much as tying primitives to primitives and then to keyword. Perhaps that's what you meant. In that sense, I agree that the amount of time and complexity of the story isn't as important, but to get to that point where you can quickly put something together and recognize when someone has constructed something you can use with/without skillful tweaking and later adjustment or replacement, I think it's important to tune into the method and get used to self-adjusting and 'mentalizing' the kanji in this particular bottom-up, radical-by-primitive way, incorporating the various tools such as SRS and muscle memory, etc., and to make the site work, sharing what we learn/create when we see fit.

To me, this 'lazy kanji' thing, basically this is what I started doing more and more just from doing 'regular' RTK properly. After doing a number of kanji, all I really needed was to spend a minute coming up with something or minimally parsing what someone else came up with (reflective of Heisig's own progressively stripped down stories but according to my own learning curve and without needing to replace them with something much more elaborate).

nest0r, that makes sense and since I'd done 600 or 700 kanji the original way before switching over I can't really say with certainty whether my way would have worked from scratch. And your correct, what I meant and didn't write clearly was linking the primitives together and then linking those to a keyword.