I did a quick search but didn't see this topic.
What do you consider a "fail" when using the SRS here on RevTK? I've been failing cards if I can't remember the story, which gives me a 70-80% retention.
But lately, due to getting demoralised and realising the goal is to remember the meaning and writing, not the story, I think I'm going to start only failing it if I get the kanji itself wrong - which I'm guessing is what most people claiming 90-95% retention do anyway.
EratiK
Member
From: Paris
Registered: 2010-07-15
Posts: 874
Yes, you need to fail the card when you get the kanji wrong, not the story.
For kanji whose stroke number is minimal (and which seem hard to remember now), the story will most likely go away sooner than later. It wouldn't be fair to fail the card when the brain is doing its work, would it?
Last edited by EratiK (2011 October 23, 5:15 pm)
amillerchip wrote:
I mean if I have to view the story at all, I'm currently failing the card. This is slowing me down no-end and resulting in a lot of time re-studying and tweaking stories. Heisig recommends that we put the story upside down on the flashcard, which I suppose is similar to here. Khatzumoto also recommends putting the story on the front of the flashcard. I think this tends to imply that needing to look at the story isn't itself cause for failing the card.
I think the main argument in favour of Heisig and Khatz's recommendation is that learning Kanji is not, strictly speaking, learning Japanese, so you should try to make it through Kanji as fast as possible, without expending more effort than absolutely necessary. That way you can move on to "real Japanese" sooner.
I have had experience with both approaches. Initially I failed the kanji whenever I was unable to recall it without looking at the story (heck, at first I didn't even write the story on the flashcards, because I didn't know that function was available on this site 
). When I started adding stories and looking at them whenever I had trouble recalling the kanji, my rate improved and the number of reviews decreased considerably, and I was able to complete the book much sooner. However, at some point I noticed that in many cases seeing the keyword gave me absolutely no hints to the story. I also completely lost the ability to distinguish between keywords that are similar in meaning without reading the story. And then when I started reading texts in real Japanese I noticed that I often have the feeling that "I know this kanji, it's in RTK 1", but can't recall anything beyond that. Because of this the process of learning the actual words on top of kanji was somewhat slowed down.
Now that I'm done with RTK 1 I have settled on a compromise that works for me. For kanji in boxes 1-6 if I see the keyword and have no idea what the kanji or the story is, I fail it right away. If I can vaguely recall parts of the story or the overall shape of the kanji, but not sure about the particulars, I take a glimpse at the beginning of the story. If I manage to recall the rest of the story and reproduce the kanji correctly, I give myself a "Yes". If I cannot recall it without reading through the whole story, I fail it. However, for kanji in boxes 7-8 if I can't recall them and reproduce them perfectly without glimpsing at the story, I fail them.
Interesting approach vonPeterhof. I like to review them all in one go though, so there'd be no way for me to easily tell which box it was in. Anyway think I'll stick with an all-or-nothing approach for now.
vix86 wrote:
The chances that I'll likely have the ability to "check the story" in real life when I need to read a kanji are slim.
This is true, but this method isn't to flawlessly remember all the kanji. It's designed to remember them all well enough to move onto further study. I'm *hoping* that learning them in context adds a lot more meaning than the abstract and frequently ambiguous keywords we are using here, and work to reinforce them in addition to the RTK method. Even with the SRS, if you keep looking at the story every time that shows up, it *should* eventually go in, right? I suppose you can use some judgement if there are problem cards you notice you're frequently struggling with.
I'm going to experiment after I've added some new kanji of passing cards more liberally (i.e. as long as I wrote the kanji correctly, whether the story was needed or not). I'll report back when I feel like I can tell if it was a good or bad idea.
kira80
Member
Registered: 2011-01-04
Posts: 23
Personally I try not to use my visual memory at all while reviewing, except when it comes to the relative positions of components. So if I can't remember the story, then I consider I can't remember the kanji, and I fail the card. I'll keep doing it that way at least until I'm done adding more cards. I'll start trusting my visual memory only when I'm sure I have the mnemonics down really well.
zigmonty
Member
From: Melbourne
Registered: 2009-06-04
Posts: 671
Betelgeuzah wrote:
/hijack
I wonder how to approach failures with my core deck. Sometimes I can't make out what a compound means before I look at the example sentence below it. Then I can deduct the meaning from the context provided. But idk if that's a reason to fail the card?
Whatever. Lots of people argue if you can understand it in context, no problem. SRS is more about getting yourself a limited minimal exposure to something, a supplement for not encountering it enough in the wild (common if you don't live in japan, are busy, etc). It's not worth taking more seriously than that.
Having said that, i mainly use a vocab deck without a sentence on the front, so i fail it if i don't know the word. My grammar deck, otoh, i pass if i know the grammar, regardless of missing the reading of a word or two.
undead_saif
Member
From: Mother Earth
Registered: 2009-01-28
Posts: 635
I think failure occurs at making the mnemonic stage, Heisig stresses the fact that you should make VIVID IMAGES of the keyword, if done correctly your retention rate should be 90%+, in my case it was 95%+.
I don't know at which frame you are, but around 1000 making mnemonic will become a habit, and also using this website's stories eases the pain in making new stories.
I've done RTK about 2 years ago, and stopped studying Japanese almost after that, now I'm relearning RTK, it' much easier to associate keywords, stories, elements and even elements' relative positions, this wasn't the case when I did it the first time.
OzarM
Member
From: Ohio
Registered: 2012-01-09
Posts: 34
I'm agonizing over this heavily all of a sudden. I'm concerned I might be progressing too fast in the wrong manner that will ultimately hurt my goal.
Currently I've been relying on the stories for most of the kanji. My routine is to write down new ones, then when I'm finished I study them and add stories, using this to practice the strokes as I do so, going through them one by one. After this I will do the reviews of any outstanding cards. If I fail anything, I study it again along with the new kanji.
The following day I study the new kanji and any failed kanji again, practicing the strokes, using stories as needed. Then late in the day I do the "final" review. For this as I go through the deck, I attempt to draw the kanji using the aforementioned link. To do this I usually end up looking up the story I couldn't recall, at which point the whole thing clicks together and I'm able to remember the strokes. The keyword alone.. not quite so often. But as long as I produce a reasonable facsimile with either keyword and/or story, I hit "Y" (I may also count it as a fail if my rendition of the kanji is particularly hideous.)
Early on as I am, I'm reluctant to use the "Easy" tag even if I can remember without the story.
Should I switch to automatic failure if I can't remember the kanji/strokes just from the keyword alone? If so, in what cases should "Easy" be reserved for?
Two more things.
One, with failed cards, I'm not sure I'm going about it correctly. Currently I just sort of lump them in with the rest of the cards that need reviewing, but they're still a "separate" review. I've noticed there's a "Add to Learned List" button. Should they be put back into the regular rotation that way? If so, as soon as possible so they come up again the following day?
Second, I know the RTK places an emphasis on keyword->kanji, but how concerned should I be if I rarely can do the reverse? Is that something that comes later, or is that a sign I'm Doing It Wrong?
