Tobberoth, we're doing self study, not testing how we do against each other. Before, I would not mark wrong if my "drop" went the wrong way or my primitives were reverse. Now I do. Before, I wouldn't mark my vocabulary cards wrong if I wrote the kanji wrong (but knew the meaning and the kanji), but now I do. As I got better, I added more to what I would consider a wrong answer.

From my understanding, you're doing japanese keyword to kanji. I'm doing English keyword plus English descriptions plus onyomi plus kunyomi to Kanji. Am I cheating? It's self study so no. Neither would you be.

Plus, let's face it: If you said "Cat - Wild dog, Flower, Field" and a guy produces a kanji that a Japanese person would associate with a cat, you've done something.

Do I recommend putting a story in the question area? Not really, but I can see how it'll benefit. Even Heisig allowed for it (write it upside down on the question side). It can be the story, just the primitives, a hint. It's up to the user to determine how much help he needs. Yes, the more help given the less certainty of how much is remembered. If it keeps a guy going then it's worth it.

@Tobberoth, in that case you're correct!

Mnemonics are aids for remembering, once you don't need it you can ignore it.
That's why I'm using a separate pdf with the stories.

As Nukemarine says, SRS is to help you remember and is valid to give yourself enough hints. Kinda like in supermemo articles, [Q] A,B,C,D,... [A] E,F,G,H >< (don't laugh if I got it wrong.) Close & graphic deletion.

For example; [Q] "Good Luck" and in my mind I have this samurai story that has nothing to do with a kanji.

Tobberoth wrote:

Going english to kanji like you isn't cheating. The english keyword doesn't tell you how to write the kanji. The story however, does. THAT'S why it's cheating.

I totally get where you're coming from here, and I do think that if relied on excessively the stories can become a crutch.

I think you're wrong, though, when you say that the the story tells you how to write the kanji. It doesn't; the words: rice seedling + soil + fat man + oven fire =heat tell me absolutely nothing about what the kanji looks like. Its just a bunch of words that recall an image. And the image helps to recall the kanji.

But in my personal experience, the kanji i have reviewed 3 or 4 times already I no longer need to look at the story. It pops into my head from just the keyword. I'm ending up at the same end goal (kanji production from keyword) as someone who doesn't include stories, I've just made the task easier for myself in the short term.

Last edited by blackmacros (2009 April 26, 11:15 pm)

mafried wrote:

blackmacros wrote:

I think you're wrong, though, when you say that the the story tells you how to write the kanji. It doesn't; the words: rice seedling + soil + fat man + oven fire =heat tell me absolutely nothing about what the kanji looks like. Its just a bunch of words that recall an image.

Wrong, it tells me everything I need to know about how to write that kanji.  Those words stand for components which I know how to write, and the more you work with the kanji in general, the more patterns you pick up.  I've been doing Heisig off and on for years, and although I've been slow in finishing I know all the primitives and have a good idea of how they work.

Fat man is typically on the right, and oven fire is always on the bottom.  Rice seedling and soil could be in a bunch of places, but given that the right and bottom are already taken, they probably combine vertically.  Therefore rice seedling + soil + fat man + oven fire MUST be 熱, even though I have never studied this kanji before (I had to look it up in the study section to make sure).  And there you have it.

Those who have been using the system for a while speak from experience when they say do not put stories in the question fields.

I have to admit you have a point there, although the point I was trying to get across is that the story (which is just a written form of the image you form in your mind) is just something you use to help remember the kanji. The end-goal is to remember and reproduce a kanji, not to remember a story. That is why the story falls away naturally while the ability to produce the kanji remains.

Regardless of that though, I've stated a couple of times that I no longer need the story for the kanji I've reviewed a number of times (3 or 4 times). So if this method is working for to cement the keyword-->kanji link in my longterm memory, whilst making it easier for me in the short term what is so bad about it?

mafried wrote:

But there's nothing special about me.  Before too long you will be at this stage too, and at that point the primitives in the question side of your cards will become a hindrance to your learning.  So my recommendation is to remove them (or stop inputting them) now before it become a problem.

But you must do whatever works best for you.

Well I plan (partly based on this discussion) to remove the stories once I have finished RtK and get into reviewing only. The whole idea behind including the stories, for me, was to make initial reviews easier. Once I'm no longer adding new kanji that will no longer be an issue and I will remove the stories. Unfortunately I've been adding the stories in the same field as the keyword which means it will be quite a task to go and individually edit each card, rather than being able to just delete a field. Oh well, I might do it as each one comes up for review.

At the moment I'm on 1550 and going at a rate of 100/day, so I should be done within the week; its not really worth changing now.

Tobberoth wrote:

blackmacros wrote:

The end-goal is to remember and reproduce a kanji

Which you won't be able to do if you need the primitives written out to remember it. The idea that the story falls away isn't really true either, it's not a good thing if you can't recall the story for a kanji. When you've trained a kanji long enough, the story shouldn't be needed to remember it but as time goes by and you're not exposed to the kanji, there might come a time when you remember 70% or so of how it looks but not the rest. At that point, you SHOULD be able to recall the story to fill in the blanks. If you forget the story completely, the risk of you not being able to reproduce the kanji increases a lot.

This is why people aren't recommended to ignore making proper stories, learning without a story = traditional learning = imperfect reproduction.

I happen to agree that including the story can potentially act as a long-term crutch. that's why I've decided to remove them upon finishing RtK. Since I'm doing 100/day, however, I have to be careful not to drown in reviews. For this purpose I think including the story is very beneficial. It makes the initial review much easier to cope with.

Even though I can recall older kanji without needing the story, I will be removing the stories to prove to myself that this is the case.

I think it is a mistake to assume, though, that just because I include the stories in my question I can't remember them. The opposite is true actually. Each time a review comes up I see the keyword and story right next to each other; each review serves to further reinforce the link between story/keyword for me. You seem to be ignoring the fact that I keep telling you that I do *not* have to rely on reading the stories after a few reviews. By that point the stories have become engrained in my head due to repetition (its not just the Answer portion of an SRS that you remember, you know).

Last edited by blackmacros (2009 April 27, 5:47 am)

Tobberoth wrote:

blackmacros wrote:

I think it is a mistake to assume, though, that just because I include the stories in my question I can't remember them. The opposite is true actually. Each time a review comes up I see the keyword and story right next to each other; each review serves to further reinforce the link between story/keyword for me. You seem to be ignoring the fact that I keep telling you that I do *not* have to rely on reading the stories after a few reviews. By that point the stories have become engrained in my head due to repetition (its not just the Answer portion of an SRS that you remember, you know).

By that argument, you wouldn't need an answer-side, just put the kanji by the keyword. We all know that's not really how it works though. It's the act of remembering it that does the trick, not seeing it.

I personally remember the story each time I see the card. You simply read it. Will you still remember the story in 5 years, even after removing it from your cards?

Again you are making assumptions about how I study. I don't simply read it. I look at the keyword, attempt to recall the story and if I can't I look at the story.

The very act of seeing the keyword and story next to each other, and reading it aloud one after the other, is bound to make my recall of the story far stronger than someone who goes straight from keyword-->kanji and only imagining the story. After all, I'm seeing the keyword, recalling and imagining the story and *then* reinforcing it by reading my story. This is of course provided that I attempt active recall of the story first before looking at it (which I do- as I keep trying to tell you).

But really, whether or not I recall the story in 5 years is completely irrelevant. Whether I remember the kanji is what matters to me. If in 5 years I am STILL having to try and remember the story to recall a kanji then something has gone wrong.

Last edited by blackmacros (2009 April 27, 6:59 am)

andresito wrote:

@blackmacros
"Again you are making assumptions about how I study. I don't simply read it. I look at the keyword, attempt to recall the story and if I can't I look at the story."

Exactly! a keyword triggers the story and the story the kanji.

If you can trigger the story in a matter of femtoseconds is almost like from keyword to kanji doing a tunneling effect, that is never being in the domain of the story... well actually is there in a negative time.

Indeed, keyword + story together is just a big question that may start going into rote memorization instead of the "imaginative memo" as the good dude Heisig calls it.

I think as long as you make a effort to actually imagine the story before (or upon) reading the story, then you can avoid that trap of turning it into a big rote memorization question. But only if you actively attempt to recall keyword-->story first, before checking the story.

Anyway this isn't really the topic of this thread and I apologize to KanjiMood for bringing it so off topic

Back to the topic at hand- I pretty easily got through my 100 kanji today. Today is a very good day for kanji. I have a class at 10, a 2 hour break for studying kanji. A class at 12 followed by another 2 hour break for kanji. Then my japanese lecture at 3. I can basically get all my reviews and most of my new kanji added before I even get home from Uni.

Tomorrow, Thursday and Friday are similarly well suited for breaking my kanji study up into chunks throughout the day.

Wednesday is going to be the hardest day to schedule myself around, so it is going to be the real test of whether fitting 100 kanji/day is feasible during a non-holiday period.

harhol wrote:

If I couldn't recall a kanji from the keyword I'd take it as a sign that my story wasn't good enough. I'm sure we've all had those stories where you think up something and it sticks in your mind permanently - that's what you should be trying to do with each kanji. Personally I'd never consider putting the primitives or even the story in the [Q] section, because if I'm even 10% unsure from keyword to kanji then I want to know about it. Including the primitives and/or story seems like a way of covering up gaps in your memory. Considering that RTK is only the first step of many, it makes sense to master it completely, because otherwise it could cause serious problems later on.

As of this discussion I have put my stories in white text to prove to myself that including them does not have a significant impact on long term recall, as long as you don't allow them to become a crutch.

My results: I have dropped from around 90% recall of previously reviewed kanji to about 80%, with most of that drop being from failing newer cards (those only reviewed maybe once or twice before). I don't have much trouble remembering cards that have been reviewed 3 or 4 times already.

I've also probably dropped to about 70% recall of newly added cards (down from about 90% with stories).

For me it is this last statistic that is important. Doing 100/day means I *cannot* afford to drown in reviews, as I've said before. Dropping to 70% recall of new items introduces a LOT of new reviews, making things a lot more difficult. If I were to do this process again (adding 100 kanji/day) I would definitely include stories initially so that you can stay ahead on new reviews. Once you finish adding, removing the stories should make sure you cement the keyword-->kanji link.

Personally I think that solution is the best of both worlds. But if you're not doing 100/day then you shouldn't have to add stories to your Q, as you won't get nearly as many reviews.

More on topic: Only managed to add 65 kanji yesterday, thanks to my experiments with stories leading to around 250 reviews (including reviewing new cards added the previous day). We'll see how I go today with adding 100, as its my busiest day at Uni.

andresito wrote:

@blackmacros,

Hey m8, how are you managing your schedule?

I'm doing grad school. Before school about 1h for "learning" only (~25 characters).
Later I will SRS in time boxes at least of 10' each. For about 30'
I do it everyday 毎日毎日

I get up a few hours before I have to start for Uni (my classes start at 9am the earliest, so this is good for me) and try to get through all my reviews. I usually can't get through everything, but I make sure I get through all reviews of old cards. Since I have around 250 reviews a day now (excluding new cards) I often can't review all of the 100 new cards I added the previous day.

Once I get back from Uni I study for about 3 to 4 hours to do my 100 kanji. Afterwards I try and get through the remaining reviews of the previous day's cards.

Any time I have left after that I study for my other subjects.

Atm Kanji>all other subjects in terms of importance.

EDIT: Ok, I managed to get 110 done today (I finished off lesson 46 and am at frame 1775)! I'm very happy with this, considering today is definitely my busiest day at Uni. While I was adding, though, over the course of the day another 180 reviews accumulated. I expect a lot more to await me tomorrow morning.

I calculated my avg cards/day since the start of my holidays 16 days ago, and it came out to 93 kanji/day Very, very happy with that considering this includes 1 week of holiday, and 1 week of Uni. So it shows that it *is* possible to maintain a 100/day pace even with other things to do.

EDIT 2: Managed to complete 128 Kanji today! Currently at frame 1904. I wonder how KanjiMood is doing? Haven't heard from him for a while.

I'm now into the home stretch and trying to get this finished off ASAP because I have a Japanese listening test on Monday at University. Unfortunately, having dedicated 100% of my efforts to learning the kanji, a listening test is about the worst thing I could have at the moment. Such is life, I guess.

Last edited by blackmacros (2009 April 30, 6:00 am)

Great blackmacros!

If you can start making a log on your post-RTK1 will be awesome.
Like how much you can understand from random readings on the internet, or how simple-difficult can you go over multilingual sites; i.e. a football site you read the news in japanese then in english.

looking forward to hear more of your success!

Ah, the piles of reviews I love them so.  I would like to say I wish I read this thread's trip down do you fail a card looking at the sentence.  I was at 100 a day up till 1700...

Normally I start out without the story on a card and try to think of the kanji, if I have any idea I'll wait up to a minute before "giving up".  Then I would glance at the sentence and if I still got it I would put it in HARD (If I was confused at all by the sentence I failed it).  This way I had 95% success rate with anki, but I was really upset because just looking at like... two words of the sentence immediately brought it to my mind, and I felt bad for even looking at it.  Hence I took the method that seems to run rampant around here of flat out failing for ever looking at the sentence.  My success rate dropped to around 75% and all of a sudden my 180 kanji a day reviews were 300, and yes, that just about killed me.  When that happened my 100/day dropped to 60/day and I ended up struggling the rest of the way to my 2042 over the next 5 days.

I don't think I would ever recommend failing cards if you knew it within like... 5 seconds of looking at sentence if you plan on doing anything like 100/day.  Now after a while, maybe... a week week and a half after finishing 100/day this might start to be a good idea, then none of the cards are "new" anymore, should all have been reviewed at least twice, and you probably have quite a bit more time to focus on reviewing those ones you felt bad about massing earlier.

If anyone is interested in why i decided to do 100/day tomorrow I'm leaving for Italy and won't be home till the 28th.  Horrible for my studies of Japanese, but... I bought an itouch and got ankimini and right now I'm sitting here going through anki writing all my sentences on my cards "hidden" so I'll be able to see it if I get stuck on a kanji on my trip.  If nothing else my vacation/class is going to give me some time to be sure I have the kanji concreted into my head before I really start moving on to other things.  Also filled my itouch with episodes of TV shows and Japanese music.  Even got Resident Evil Extinction in Japanese lol.  Probably going to throw a bunch of Tae Kim sentences on another anki deck and get it uploaded to ankimini too so I can feel like I've started the sentences portion.

He did say "hidden". Sure adding the story to the front of the cards makes going through the kanji quicker.  Does it teach you how to recall the kanji well without having that aid though?  If I recall, there was at least one member who said he had a hard time recalling without looking at the story.

Last edited by vosmiura (2009 May 09, 10:39 pm)