asriel, that really helped me! I feel so stupid that I didn't even notice that huge table and what it meant... thanks : P

I'm just wondering, if I want to have that word lookup thing, and sound, and furigana etc, I'll need to download all those plugins right? They're .sh files right, so do I download them and then will they be plugins I can access by opening them through anki? And with the sound files, if I go with the "remote" ones does that mean I don't need to download, but it's accessed through the internet?

thanks : )

Word Lookup --> I'm afraid I have no clue

Sound --> Should be built in (?). Remote means that you'll be playing them from the server every time. Local would be saving them to your computer and playing them there (like if your internet dies, you put it on your iPod, etc...)

Furigana --> Yes, plugin. Get it through "Download Shared Plugin"(?) in the menu. I think it's built into the "Japanese Support" plugin, but you'll need an addon to get it to work in every field. Otherwise it just works on the field called "Reading."
I'm not 100% on this Furigana thing, but I do know that it is a plugin, and the "Japanese Support" one is involved.

Thanks so much for all your help. I now know what to actually do to turn this thing into a deck. I really appreciate that you have helped me with this : )

hey, sorry to keep asking questions (this one's not directed at you asriel because you said you yourself aren't 100% sure about furigana addon), but when I used the xxfurigana addon with the command <!--fon-->, it comes up with something like this:

class="fm51de892a89ecea03">漢字

Now, the furigana reading for the two kanji, in this example ”かんじ”, appears over the top of that whole string I just typed above. I don't know why this "class=" thing is popping up. Is it something wrong with how I typed the code in? This is how I wrote the answer segment for my cards, which has the furigana:

<!--fon-->%(vocab-furigana)s <br>
%(vocab-meaning)s <br>
%(vocab-pos)s <br>
<!--fon-->%(sentence-furigana)s <br>
%(sentence-meaning)s <br>

The field names here correspond to those on the anki page for the Kore deck.

Sorry to keep asking questions, I think this should be the last one for a while : P

Same thing is happening to me, and I have no idea.
I PM'd or emailed (not sure how it works here) cangy about it, because I think he's the one who made xxfurigana...

I have no idea why it's happening, but I would love to find out as well.

vinniram Wrote:class="fm51de892a89ecea03">漢字

vinniram Wrote:<!--fon-->%(vocab-furigana)s <br>
%(vocab-meaning)s <br>
%(vocab-pos)s <br>
<!--fon-->%(sentence-furigana)s <br>
%(sentence-meaning)s <br>

I use <!--fon-->text<!--foff--> in my templates to close it properly. If I remove <!--foff--> I get a similar problem. It screws up the card formatting.

Edit: The lack of the above doesn't appear to necessarily break anything (although it does in my case, specifically where the sound string is placed)

In your card fields, do you have a space before the rubified word? If not, this also breaks the formatting. E.g. the field must be " 漢字[かんじ]" and not "漢字[かんじ]" although this should be correct by default if you import the spreadsheet.

Color formatting inside furigana fields usually also result in quirky behaviour unless you color the entire "ruby word".

This is just what I've encountered while messing around. Other than that, I wouldn't know.

pm215 Wrote:I've been using this deck with word-in-kanji-only on the front and word/sentence/audio on the back. However this is because I'm basically using it to fill in things I don't know or half-know -- about 60% of new cards I suspend immediately as already-known, and I'm deliberately trying to catch the words I'd understand in a sentence but don't quite grasp enough to know if I see them out of context. I'm not sure I'd approach it this way if it was all unknown vocab to me (my other deck of random encountered vocab is sentences-on-front-of-card.)

Results? Well, I'm now just over halfway through it (at 20 new cards plus ~30 suspends a day) and I definitely see these words in other reading I'm doing. So I'm happy with it. I think the main benefit for me has been that it's a premade deck so it's much easier to sustain a regular pace of newly learned cards than if I have to find and add sentences myself. Plus having the audio is nice.

Thanks for your feedback. Taking this into account plus reading other threads, it seems most people switch to vocab after maybe 2000-3000 sentences. I think I'll just stick with the sentences for now then.

ah yeah what you said about there needing to be a space before the rubified word is true. however, even though I imported from spreadsheet, rubified words at the beginning have no space before them obviously, so I guess I"ll have to go through and manually add spaces :'(

This idea with the spaces is pretty interesting and I'm going to have to take a look at it.

Would it work if, in the card template instead of the individual cards, you added a space?
ala: "<!--fon-->_____%(RubyWord)s<!--foff-->"
Space exaggerated by the use of _____

??

Yeah, the problem with xxfurigana comes from the plugin (and the original furigana support, too, afaik) expecting a space before a 漢字. (Which is especially annoying if you have a plugin that removes leading spaces, like I do, or like someone mentioned, use HTML tags.)

The following version should fix this by also allowing a ">" before 漢字. It's a bit stupid and there probably should be a better way, but it works without issues for me.

Replace the xxfurigana file in your plugin folder with this one:
http://gist.github.com/540764

I couldn't get the above to work, but what I did was to put a space at the start of each field using Anki's search and replace. Problem solved! Thanks for the suggestion about the spaces.

In the browser, select the card you want to change (probably all of them), then go to the edit menu and then find and replace. Make sure you limit the operation to the desired field 'Reading' or 'Kana' as appropriate, then in Find: put "(.*)" without the quotes, and in Replace with: put " \1" again without the quotes. Next tick the regular expression box. Do a backup beforehand and do remember the edit undo in the browser menu. I just posted this in case anyone was trying to add the spaces manually.

Actually, my formatting is still not quite right - I have all the furigana showing properly except that the last one in the sentence is misplaced - floating above the others. That may be a separate problem though ? or something specific to the deck I was testing on ??

thanks rachels!!

I'm happy I held off doing manual spaces for a little while. Your fix has worked perfectly : )

I haven't yet encountered the problem you have been having, but if I do I'll let you know.

Hi all. I got the idea to sort the sentences by word frequency, in the same kind of way as the current kanji sorting options.

I don't really know what I'm doing with MeCab because I can't read the documentation ;). The result looks generally plausible though.

Version 2: Where the frequency of the rarest word in the sentence is equal, sort on the second-rarest, and so on. Words that MeCab didn't recognise are assumed to be unique (count of 1).
http://s000.tinyupload.com/index.php?fil...1544192924

Version 1: Sorted only on the rarest word in the sentence, so after the first 2580 sentences, the sort order is arbitrary. Words that MeCab didn't recognise are coming up as high frequency.
http://s000.tinyupload.com/index.php?fil...8261075107

The line number is the kore "index" field.
The first column is the sentence broken into a comma-separated list of words (and not-exactly-words) in dictionary form.
The second column is the new sort order.
The third column functions in a similar way to the freq-kanji field etc. Though the number is counting the number of times this item appears in the kore sentences.

To merge with a kore spreadsheet, first sort the spreadsheet by "index" field (column P on the unedited text file) then paste in the new columns. If you have deleted any rows, it will be more complicated.

If other people are interested in this, I can try to improve it, document what I did and make my scripts presentable.

I am currently using the Version 2 sort order myself, and will report back after a while about whether it was useful in practice.

(Post revised heavily to make it less confusing.)

100 sentences in, I am liking this order so far. As I intended, it's giving a lot of repetition of the words in different contexts, and this definitely helps me remember them.

I was struggling with the grammar at first but it seems to be OK now.

I only tested my order up to 150, then started experimenting with other ways of selecting the next ~30 sentences. Maybe I spend too much time doing that and not enough time learning the sentences, but programming practice is important too ;)

Hey I'm just wondering how many kanji from rtk are covered by core2k and 6k combined. Is it all 2042 or is it only a few. Thanks if you can be able to tell me this one.

vinniram Wrote:Hey I'm just wondering how many kanji from rtk are covered by core2k and 6k combined. Is it all 2042 or is it only a few. Thanks if you can be able to tell me this one.

vocab: 1510 rtk1, 96 rtk3, 2 other
sentences: 1545 rtk1, 103 rtk3, 2 other

cangy Wrote:vocab: 1510 rtk1, 96 rtk3, 2 other
sentences: 1545 rtk1, 103 rtk3, 2 other

is there a community consensus about the best way (most efficient, least amount of overhead time with deck creation) to finish of readings of the remaining joyo kanji not covered in the kore series?

I am messing around with this but I am confused what the #lookup-sentences is supposed to do? I tried to run it through the .sh script but I think it failed=/ What is supposed to happen after you runned it through the script? also is the lookup-sentence an important feature? or is it not necessary?

also I placed the audio of the vocab and sentence both on the back side. But the sounds play right after eachother. Is there a way to stop that?

thanks

kodorakun Wrote:

cangy Wrote:vocab: 1510 rtk1, 96 rtk3, 2 other
sentences: 1545 rtk1, 103 rtk3, 2 other

is there a community consensus about the best way (most efficient, least amount of overhead time with deck creation) to finish of readings of the remaining joyo kanji not covered in the kore series?

Honestly, by the time you finish the core/kore series, you're in a great place to just start reading. Read, read, read and read. And then read more.

If you _need_ readings for a particular reason, ala, JLPT or its ilk, then simply grab a deck designed with that vocabulary in mind and study towards it.

But, for general purposes, by the time you're done with that kore deck you're going to have a foundation solid enough to just start reading and the rest will show up. I'd probably just start a pure vocab deck at that point, dropping in new readings you were unfamiliar with and had to look up, etc.

what sort of level of grammar is covered by kore? is it JLPT N1-level grammar, or simpler than that. Because I'm considering whether to, at some point, work through and sentence mine the dictionary of Japanese series, either before or after Kore, except I'm realizing it'll take me a few years... however, I've heard that kanzen master isn't as comprehensive and I should avoid it, although that was just a couple of people telling me that.

vinniram Wrote:what sort of level of grammar is covered by kore? is it JLPT N1-level grammar, or simpler than that.

It doesn't cover grammar at all; the aim of Core is teaching vocabulary, not grammar. So any grammar it includes is just incidental to providing an example use of the vocab. You won't get: grammar in a sensible order; comprehensive coverage; sentences which are helpfully representative of the grammar point; explanations of what the grammar actually means.

Quote:Because I'm considering whether to, at some point, work through and sentence mine the dictionary of Japanese series

IMHO the "work through" bit is more important than the "sentence mine" bit. All this grammar at least up to JLPT2 and some of JLPT1 is pretty common, so you're going to get the repetition and reinforcement naturally as you read, *provided that you understood the grammar in the first place*. I'm unconvinced that just throwing all the sentences from the D*JG books into Anki will be very effective.

vinniram Wrote:I've heard that kanzen master isn't as comprehensive and I should avoid it, although that was just a couple of people telling me that.

KM probably covers a wider range of grammar points, but it has approximately no actual explanation of the grammar point; so it's broader but much less deep. Handy if you want revision or JLPT-test-style quiz questions, but not a primary reference.

loonytik Wrote:I am messing around with this but I am confused what the #lookup-sentences is supposed to do? I tried to run it through the .sh script but I think it failed=/ What is supposed to happen after you runned it through the script? also is the lookup-sentence an important feature? or is it not necessary?

it's already included so you don't have to do anything -- the link is just there for reference -- see the fields example for what it contains, but basically it just saves you having to look words up manually (shows transitive/intransitive verbs etc.) and helps in separating the words in the sentence if you have trouble with that

loonytik Wrote:also I placed the audio of the vocab and sentence both on the back side. But the sounds play right after eachother. Is there a way to stop that?

you could add a silent track to your template between the other two

vinniram Wrote:what sort of level of grammar is covered by kore? is it JLPT N1-level grammar, or simpler than that. Because I'm considering whether to, at some point, work through and sentence mine the dictionary of Japanese series, either before or after Kore, except I'm realizing it'll take me a few years... however, I've heard that kanzen master isn't as comprehensive and I should avoid it, although that was just a couple of people telling me that.

KM1 and KM2 are widely praised on this forum, as are the Dof(B, I, A)JG. What I do is use the KM2 order and sentences, but supplement with DofIG (sometimes beginner) to get a full understanding. So far it's a pretty good combination.

Like was said, the Core series (and KO2001 book 1 and 2) are vocabulary resources so the grammar is kept simple is the sample sentences for the most part. Likewise, grammar resources like Tae Kim, Kanzen Master series and Dof**JG books use simpler vocabulary in their sample sentences.

As such, while I highly recommend doing both vocabulary and grammar resources, one should sentence mine actual Japanese sources (subs2srs, manga, news, books, etc) to see what it's like having both native words and grammar used in a wide range of combinations.

With the km series , about how much of the content from the do*jg series does it cover . By this i mean to Say, Does km give example sentences for most of the various things covered in the dictionary series ?