Yeah that's what I thought. I've never seen a site as professionally constructed as this that wasn't at least trying to break even. It's a good service though, I would probably pay for more of the same.

Even if it was fully intended to stay a free website, I doubt they would want people coming in, stripping out all the sentences/audio/images to put in their own SRS programs, and then leaving. There's a whole community thing and inbuilt learning apps. That's why I'm a little wary of going directly to them, in case they decide to change the way the sentences are formatted so they're not as easy to strip-mine.

Semi-related, but if anyone doesn't know how to do the process I mentioned, I could upload the two Steps (3+4) I've done already. Does anyone know if that would be okay to post here?

[ファブリス]

shakkun Wrote:Does anyone know if that would be okay to post here

If you're talking about data extracted from the iKnow website, it is not appropriate to post links here. That shouldn't stop interested members here to get in touch with you via your profile to share tricks for collecting sentences though.

shakkun Wrote:And the images all have title tags which correspond exactly to the sentences they belong to.

How'd you go about solving this. I have images with no clean way to associate them.

kazelee Wrote:

shakkun Wrote:And the images all have title tags which correspond exactly to the sentences they belong to.

How'd you go about solving this. I have images with no clean way to associate them.

I'm not sure I understand this question, but if you use DownThemAll (plugin for firefox) you can download all the images on a page. Set "Renaming mask" to *text*.*ext* and it will save the images in the format "彼は国会議員です。.jpg" etc. This way you can just duplicate the sentence column in Excel and do the following find and replace in EditPad Pro:

Find: \n
Replace: .jpg">\n<img src="

When I get a chance I'll put the ones I've already done on the new AJATT forum. I'm not good at explaining these things so I might just do the Steps 4-10 and put those up too. Including grabbing all the media it probably takes 15 minutes each.

Ha. I was searching for that plugin but couldn't rememeber the name. I was using flashgot up till now. Thanx.


Edit: Oh poo, for some reason all I get is 0 for the file name. I crack... for tonight that is...

Okay I put steps 3-10 on the AJATT forum, under Copyrighted Works so you at least need an account to see the post. I didn't do 1 and 2 because I'm not using them, but if you can see the format it might make it clearer. If for whatever reason people want the first two and really don't know how to format the pages, I can do those too (or try and explain a little better?)

Ahaha. Just when I was getting the hang of this RegExpression thing.

Edit: I was trying to make a macro that did everything for me all at once. All the macros disappeared once I closed the program, though.

Shakkun, I have to ask which AJATT forum. They don't seem to be on the two that I'm aware of.

Oops, the ajatt.pseudosphere.net one! Sorry I should have made that clearer. The Copyrighted Works section is hidden unless you make an account and log in.

kazelee Wrote:Ahaha. Just when I was getting the hang of this RegExpression thing.

Edit: I was trying to make a macro that did everything for me all at once. All the macros disappeared once I closed the program, though.

Really, are you using EditPad? Never happened to me. Are you using the free version maybe? One thing I like about it is you can save your favourite search and replaces. Saved me some time. Anyway, if you do the earlier ones you should upload them to that post on the forum too. Also let me know if there's something you can't figure out.

Yup I got the EditPad.

I assume it has something to do with the way I closed the program because the ones I made a few hours ago are still there. :oops:

BTW

Thanks for all your help!!!

Shakkun, could I make the request then for you to do Core 1 and 2 the same way?　Thanks also for all the help.

No problem. I've added them to the same Mediafire folder as the others.

I post this words to thank shakkun for making it easily possible to import this to Anki! Thank you!

My method of importing it to Anki is by saving the excel file to a txt tabbing. However, when importing it adds " before and after the <> and the Japanese text. So I have to manually remove it if I want Anki to show the picture.

<img scr="ゲゲゲゲ。.jpg"> becomes "<img scr=""ゲゲゲゲ。.jpg"">"

Am I doing something wrong or is there a simpler way to export it to Anki?
I don't mind that much to remove it manually as I shouldn't complain to have them already in excel but if there is a way, I would be happy to know how!

Are you exporting directly from Excel or copying and pasting to a txt document? I actually use OpenOffice Calc so I can't comment on that. This is what I do though:

Select the contents of the spreadsheet and copy
Paste into Notepad
Save as UTF8
Import text file into Anki

If you do that and it does the same thing, something is going wrong with the Anki import. Are you using the latest version? Is this happening to anyone else?

shakkun Wrote:Are you exporting directly from Excel or copying and pasting to a txt document? I actually use OpenOffice Calc so I can't comment on that. This is what I do though:

Select the contents of the spreadsheet and copy
Paste into Notepad
Save as UTF8
Import text file into Anki

If you do that and it does the same thing, something is going wrong with the Anki import. Are you using the latest version? Is this happening to anyone else?

Argh, I did "Save As" in Excel and I chose the txt extension.

I tried copying and pasting directly to Notepad as you said and it works perfectly
Why I didn't think at that :/ Copy/Paste for the win.. ... "Save As" for the lose

Additional changes and upcoming changes from iknow:

1. Addition of a Kanji intensive studying method. Not sure how that works.
2. The word list that the sentences are based are themselves derived from that "Most Common" list developed from scanning newspapers. Not sure if this is a good thing, but it's better than a random list. When you start Unit 3, the vocabulary will get noticeably harder.

1. Other languages will be added, Chinese and Korean look to be next year.
2. There will be a kana training/practice section.
3. More advanced lists of vocabulary will be added. Seeing that they used TOEIC lists for the English training, they might use the JLPT lists for the Japanese section.
4. More intuitive ways to develop user lists.

Plus, the journal section is not too bad. Each of my Japanese posts have been corrected by native speakers.

Unfortunately, I've placed iknow ahead of my Anki reviews, so I'm sure I'm upto 200 due in Anki. However, I know I'll catch back up in Anki, finish up UBJG and move onto either KO2001 or iknow sentences.

shakkun Wrote:Are you exporting directly from Excel or copying and pasting to a txt document? I actually use OpenOffice Calc so I can't comment on that. This is what I do though:

Select the contents of the spreadsheet and copy
Paste into Notepad
Save as UTF8
Import text file into Anki

If you do that and it does the same thing, something is going wrong with the Anki import. Are you using the latest version? Is this happening to anyone else?

Ugh. Your directions are crystal clear...but I am having major issues with having Anki processing my imported TXT file properly.

For my first (and only) card that appears after importing, I get "iKnow<br>Japanese Core Step 1" on one side and "http://www.iknow.co.jp/list/19053-japanese-core-2000-step-1" on the other. The import log also indicates many duplicate expression errors.

Do you have to set up any special fields or settings in Anki for the new deck?

I am using the latest version of Anki (0.9.8.1) on Vista.

Thanks for putting the effort into setting this up for all of us! I just wish I could get my deck to work, as it seems like a really simple process.

Yeah, you need to specify the field mapping when you import. You'll also need to add a few fields to Anki's default Japanese model.

1) Under Model Properties>Fields, add fields for Audio and Image. You'll probably want to untick the Required? and Unique? boxes.
2) Under Model Properties>Cards, change the card model to look like this. This will test you on kanji sentence + image to kana reading + english translation + audio pronunciation. If you want something different you can just move the fields around or add another card type.
3) When you import, copy the field mapping in this image. You can just discard the first two fields. I include them because I include the source/URL of every sentence in my deck. If you want that, you'll need to add two more fields: URL, and Source. It's not necessary though.

I also would suggest adding a 5 digit code for each of the sentences if you're importing. Something like 01-138: That'll be Sentence batch 01 (of 10 in this case), sentence 138.

It's probably not necessary, but something to quickly organize a list is useful to have in my opinion.

@ shakkun & nukemarine:

Thanks for the Anki instructions & advice...I'll try it out after work later this afternoon. I had a feeling it was due to my lack of experience importing "decks"...I typically typed in my sentences & import audio from scratch, but I am quickly realizing that making my own sentences is eating far more time than actual studying. Not a good thing.

I must say, I absolutely love what the iKnow site is doing for my vocabulary retention so far. My tentative plan is to complete a lesson step to 100%, and then at that point start importing those sentences into Anki while I move on to a new step. We'll see how that works out long term, but this pattern is fitting nicely into my work schedule as iKnow study sessions are centered around 10 word blocks that I can easily fit into a break or lunch through out the course of the day.

It's been about two weeks (well, two weeks if you add the days together that I studied). What I've been doing:

1. Study roughly 1 hour all together.
2. Do 10 items selection.
3. Do ONLY 30 items per Step per day (as recommended).
4. If by doing 30 items, I'm under 1 hour, I go to the next step and start on that one.

What I've been noticing: Initial studying of 10 new items currently takes about 15 minutes. After that, it takes about 5 minutes to review 10 items that have previously been studied. A word that's reviewed successfully (no mistakes) gains about 25%. It seems a word pops up for review about once a week, so after 4 weeks you should be done with that word assuming no mistakes. With this, I seem to be adding about 20 new words a day into the review process.

How this is affecting my Japanese: Ok, I realize that not increasing my vocabulary affected my passive studying (TV, Manga, Music, etc.). Now I'm getting fresh words at a decent pace. Surprise, surprise, that these words pop up in my passive input making that more fun.

My current plan: As BigAmish is thinking, begin putting these sentences into Anki. There's little point to doing it while iKnow is reviewing it, but definitely after a word is 100% I'll put that sentence into Anki with a generous initial review interval (probably 20 to 30 day spacing for normal).

Success!

I had to fiddle with media file locations for things to work properly (again, my lack of experience with Anki's inner-workings), but everything works great now.

Thanks again! Between these 2000 sentences, and the 8000 that are supposed to be coming this month, I should be good to go for a while. The only other sentences I plan on SRSing are either out of specialized interest, slangy catch-phrases, or will be grammar-specific. Sweet!

BigAmish Wrote:Success!

I had to fiddle with media file locations for things to work properly (again, my lack of experience with Anki's inner-workings), but everything works great now.

Thanks again! Between these 2000 sentences, and the 8000 that are supposed to be coming this month, I should be good to go for a while. The only other sentences I plan on SRSing are either out of specialized interest, slangy catch-phrases, or will be grammar-specific. Sweet!

Really? 8000 more sentences on iKnow?
I haven't heard of that! Where have you seen this? ^^

This sounds like it will be awesome when the intermediate series comes out.

wonk2001 Wrote:

BigAmish Wrote:Success!

I had to fiddle with media file locations for things to work properly (again, my lack of experience with Anki's inner-workings), but everything works great now.

Thanks again! Between these 2000 sentences, and the 8000 that are supposed to be coming this month, I should be good to go for a while. The only other sentences I plan on SRSing are either out of specialized interest, slangy catch-phrases, or will be grammar-specific. Sweet!

Really? 8000 more sentences on iKnow?
I haven't heard of that! Where have you seen this? ^^

Yeah, I was basing the 8000 number on their total estimated number of sentences when everything is said & done. Sorry about that! I assumed that both intermediate & advanced sentences were coming *this* month (thanks nestOr for the link), but even still...we will have a whole lot of sentences to chew on relatively soon. Good news for the AJATT folks!