Hi Everyone,

I don't know if this has been mentioned before...or maybe it's a stupid idea...but it occurred to me after checking out the KO2001 course that if you work with every single sentance for each reading of the kanji, you'll end up reviewing for example roughly 1500 sentances (based on just level 1 which contains 555 kanji and based on an average of 3 readings per kanji).

However, since some of the sample sentences contain 8-9 kanji, if it's possible to select key sentences that do not repeat any of the kanji reading, you could actually only review about 200 key sentences.

Now, you're not going to have repeated kanji but i feel that the repeated kanji is not needed as much when using anki.

I was thinking about starting KO2001 and I'm primarily interested in studying the first 1100 or so Kanji before starting to hit native material - studying 400 sentences is a lot more appealing then 3000

What do guys think...good idea? or isn't just another shortcut ideas that don't don't really make sense.

John

From my experience: I would wind up failing those kinds of sentences constantly, because you'll have too many fail points in each sentence. 8 new vocab words = 8 things that can go wrong. 1 or 2 new vocab words = only 1 or 2 things to go wrong.

Also, I've never learned a vocab word just from reading it in one sentence. It usually takes 3 different "exposures" minimum to get it in there for me and improve my overall retention. So I use 3 shorter sentences, rather than the long, convoluted KO sentences. (I used them for the first book, then switched at book two.)

Also, long sentences = longer time spent reading = longer time spent reviewing.

Shorter sentences will also give you a better chance to see what's not sticking, and push the stuff you get right away off to the distant future.

All IMO. I'm sure there's other input.

What you'll find out is that it's better to have one card per word you're learning. It's not so much about the kanji, but the word using that kanji. For a sentence with multiple words (new ones at that) per sentence, you might find it better to duplicate that sentence for each word.

In other words, have a card per word with the sentence there to supplement learning the word in context. So, that means 3000 sentences (lotsa duplicates) as I think that's how many words are taught.

I agree except for the usual, I think thinking in terms of fail points and reducing them is a bad idea. i+X rather than i+1.

Just be flexible in grading and when possible make the linguistic components complement one another (which often might entail less new vocabulary items but might also entail more since sometimes they reinforce one another), so that the ratio of information in the initial encoding process is about using the levels of processing effect to enhance encoding and recall, rather than that classic effect of trying to memorize too much and ending up with less. In the long term the number of 'fail points' or rather 'learning points' is non-holistic and always dynamic and becomes more minimal as you primarily focus, per review, on what you previously failed. Regardless of a single card being repeated, you're algorithmically internalizing only new or partially learned information according to self-grading, while reinforcing the better remembered/passed information.

Sorry for the ramble, the phrase 'fail points' immediately inspires a reflexive response in me.

Sorry if my glass is half-empty.

From my experience, knowing the readings doesn't help that much. Maybe before the age of computers, when we had to use our dictionaries and search by hand to understand what something means... but now we have smartphones and computers with radical lookup, we can copy-paste text from the internet, etc.

Words need multiple contexts to grasp the entire meaning... and even then, even if you did all of KO2001 sentences, you will not understand the nuances and usage behind certain words. You'll need either specialized studying of that specific word or lots of indirect studying through mass exposure.

If you're gonna do the sentence method, you're going to do a lot of sentences anyway, regardless of the source. You say that 3000 sentences isn't appealing, but why not? Eventually, you're going to study more than 3000 sentences in your Japanese studies and you're going to put in a lot of time studying anyway. Might as well learn them now.

rich_f Wrote:From my experience: I would wind up failing those kinds of sentences constantly, because you'll have too many fail points in each sentence. 8 new vocab words = 8 things that can go wrong. 1 or 2 new vocab words = only 1 or 2 things to go wrong.

Also, I've never learned a vocab word just from reading it in one sentence. It usually takes 3 different "exposures" minimum to get it in there for me and improve my overall retention. So I use 3 shorter sentences, rather than the long, convoluted KO sentences. (I used them for the first book, then switched at book two.)

Also, long sentences = longer time spent reading = longer time spent reviewing.

Shorter sentences will also give you a better chance to see what's not sticking, and push the stuff you get right away off to the distant future.

All IMO. I'm sure there's other input.

I must say that althought I am a strong advocater of KO2001 to get started , on the middle term it's clearly unefficient .
1- too long . Sometimes way too long ( Kanji In Context is worse)
back when Katz wasn't rambling and repeating himself , he clearly pointed out to KISS your sentence . Keep It Simple Stupid . Well known principle extremely relevant in language learning .
too many failing point
too long to review
2- not enough context in the card / no fun : Devotion and strong will only carry you so far.... There's no fun in reading those abstract sentences ad vitam eternam
3- not enough context for the word . I have to go with rich_f : except if the word makes a very strong impression on you (I think I learned ぼったくり 　the very first time I heard it ) you need to see it several times in different context to stick , not to mention being able to use it properly .

KO2001 is a crutch like RTK . Use it properly (which means no shortcut) and then get rid of it .

Length isn't an issue because you decide what to focus on, so 'keeping it simple' advice is incorrect in that sense of 'quantifying' information (but minimizing overhead should be a focus if you want to interpret 'simple' that way, which includes a production line approach as well as awareness of how more elaborate complementary information actually makes memorization easier), but the construction of sentences in KO2001 and their dryness makes them useful more as a supplement. I just suspended that KO2001 deck w/ audio and unsuspend cards as a secondary general/fixed corpus (the primary general/fixed corpus, where I unsuspend cards and work on subvocalization and writing and vocabulary being the Core 6000 deck) to hear it and see it used in another context, only grading the cards as single vocabulary items. The plethora of terms in rich grammatical contexts and the clear native audio make these sentences 'deep' as a corpus, methinks, for both kanji and vocabulary, so that singular limited focus as a secondary corpus gives it a long shelf life as sentences always have something to offer.

Interesting discussion...

Sounds like there are some 'hidden' benefits for doing the sentences as is...

Thanks for the feedback.