Joined: Jan 2013
Posts: 151
Thanks:
0
Do you guys think core 2k/6k/10k is better than sentence mining?
For example, Japanese Level Up recommends doing 1000 sentences from Genki I and Genki II. You build on the previous knowledge you know and provide as little english as possible. Then by 1000 cards you switch to J-J only. Once again you build on what you know adding sentences where you recognize all the words except one or two new words. I have been reading a lot about the sentences because I finish RTK in a couple days. I could imagine the only reason core would be better is because it has audio in it.
Also with tae kim I was thinking about putting the whole sentence on the front of the card, all in japanese. On the back put the word that is new with the meaning of the new word Example. Front: 魚だ。 Back: だ (is). Another example Front: 魚じゃない。 Back: じゃない (is not). Over time with my sentence deck and tae kim deck sentences would be much more complex. Several other guides I've been referring to, stress input before output. Would this be an ok way to approach tae kim's grammar?
Joined: Jan 2013
Posts: 151
Thanks:
0
One more thing, I'm previewing core to see if I like it, how exactly do you get the audio to work? I'm sure it's simple but I can't get it to work for the life of me..
Place the audio files directly in the "collection.media" folder, usually found somewhere like user/documents/anki. As long as the file names in both the card field and the collection.media folder match up, they should be automatically played upon review. By default, you should be able to press the "r" key to repeat audio.
I mean no offense but this information is found both on the first page of this thread and by searching for "core2k 6k audio" in the search engine on this website. Getting audio to work is also outlined somewhere within the Anki user manual, found on the Anki website.
Joined: Feb 2011
Posts: 18
Thanks:
0
The new card order for the RTK deck is sorted by Heisig number, not 2K1 Odyssey. I can't figure out a way to change new card order. Any advice?
Joined: Jul 2007
Posts: 2,313
Thanks:
22
killazys,
You pretty much have to re-import the deck from a sorted spreadsheet to get it to show in 2k1 order. However, the deck is meant to be learned in RTK order. You're just shown only those kanji that are in the 555 kanji from the first book of 2k1 Odyssey. 2k1 Odyssey order is a great for learning vocabulary words, so the vocab decks based on frequency are sorted in 2k1 order.
To summarize: 2k1 tells you which kanji you should learn first, while RTK tells you the order. Frequency lists determine which vocabulary words you should learn first, while 2k1 is used to organize that list.
Joined: Feb 2011
Posts: 18
Thanks:
0
Nukemarine,
Thanks so much! Especially for going through all the effort and giving this guide to the entire community..
Joined: Apr 2009
Posts: 128
Thanks:
0
Coming off RTK already, I'm not sure "optimized" is the best way forward - has the deck order just been changed? Today I had to learn evening, lunch, breakfast, dinner, supper in this order. Lol these words are ultra similar and its just impossible to remember properly. If they're broken up in a different order in the original core I would of preferred that.
Joined: May 2011
Posts: 51
Thanks:
0
I used "Opt-Voc-Index" as well and ignored the instructions in the OP about suspending/un-suspending, just went in anki order. I believe those were only applicable to the older non-optimized deck.
Going through it this way though, for me I like the optimization. It makes it easier to learn the readings when the same ones are used in different words close to one another. Sure there are many that have similar meanings but that was all over the place in RTK too and at least here you have the pictures to help you differentiate. Don't know what I will do after 2k and I have no more pictures though...I'm very much a visual learner and they help a lot.
Only other downside I see to the optimized order is you really aren't learning by frequency. There are N1, N2 words mingled in with N5 words. But like RTK, if you have committed to going through the whole 2k/6k/10k this shouldn't be a problem.
Edited: 2013-08-16, 9:23 am
Joined: Jul 2007
Posts: 2,313
Thanks:
22
Yes, use the "Opt-Voc-Index".
@kanjimood, I understand you see it as a problem, but others see it as a benefit. There's benefit in grouped meanings to learn words in mass. That's why many textbooks teach vocabulary based on rooms of the house or job locations. With optimized core, you get similar words put together to learn which forces you to realize their similarities and differences. This can be beneficial when it comes to active and passive verbs, days of the week on top of noticing kanji's onyomi and kunyomi.
Now, if you are having a problem, there's no rule against suspending a card. If you think too many of a similar type are being introduced, just suspend the extras and learn them a little later.
@Meteki99, technically you are learning by frequency but in bulk groups instead of individual order. In this case, the frequency is based off of Japanese newspaper articles as it was the most readily available at the time iKnow created Core 2000. As such, its a different metric to whatever the powers that be used to determine what words go into the N5, N4 ... N1 lists. Regardless, the list does use frequency to get a group of words that are then sorted.
Assume there are 50,000 words. The list would take the top 4% (2000 words) and sort those in an easier to learn order. It then takes the next 4% and repeats the process. Its an easier matter to use smaller groups. True, I fudged with a bit using the KO2k1 book 1 to limit which words are first learned but the impact should have been minimal. A few common words like 大丈夫 are impacted due to using an less common kanji for instance.
Joined: May 2011
Posts: 51
Thanks:
0
I see what you are saying about the bulk groups.
When I hit 1000 on the optimized index I threw everything in a spreadsheet and highlighted the ones I knew and started comparing the optimized index to the default index. All the cards <2K in the optimized index are also all <2K in the default index. So it is like when the index was being optimized, the bulk group of the first 2K was reordered for ease of learning.
I didn't see any patterns for 2k~6k but wasn't looking closer either. I couldn't tell at the time if the next 2k were considered a group or just the remaining 4k was optimized as a whole.
Joined: May 2011
Posts: 51
Thanks:
0
Anyone else find the Tae Kim deck pretty vague at times with what it presents you on the front of the card (just the sentence)? For each card my thought process is figuring out what grammar point is being illustrated, to recalling the grammar point, and reformatting it so the answer fits in the sentence.
After getting through 200 essential sentences often I find myself having trouble figuring out what grammar point is even being illustrated in the card. In my head, looking at just the sentence, there seem to be many different acceptable answers that could be used. For some, I have resorted to just memorizing the sentence to recall the missing piece, but this seems silly in a deck that is supposed to be reinforcing grammar.
I have gone through and added a field to the first 50 essential sentences that contains the section header from which the card originated, to see if that helps. Did anyone else run into this?
Edited: 2013-09-04, 12:59 pm
Joined: Dec 2007
Posts: 240
Thanks:
0
You can also change the color of certain words in the sentence, if you want to mark a specific grammar point.
Nukemarine (and/or others) I have a stupid question in relation to the optimized japanese vocabulary 2k/6k deck I'm using: when people talk about "production cards", is this essentially the same thing? That is, the vocabulary item in English and having to recall the correct Japanese equivalent?
For whatever reason the thought came to mind during reviews today.
Joined: Jul 2007
Posts: 2,313
Thanks:
22
uisukii,
Yes, by production I meant English word to Japanese word. Recognition is Japanese word to meaning/concept.
Thanks for the speedy response.
Without delving too much into the whole production/recognition debate- would it be fair to say that how you have set up this deck may very well possibly provide the framework for a deeper level of comprehension than as per what you've described as recognition?
Either way, I must say that the way in which you've optimised this deck is simply outstanding for learning. Does remind me of RTK.