Background: http://forum.koohii.com/showthread.php?p...#pid180163

There is an Anki deck with all the sentences from the great Basic/Intermediate/Advanced Guide to Japanese Grammar series. The problem is that it contains typing mistakes, inappropriate kanji usage and wrong autogenerated readings.

This is a group project to clean up and improve the deck.

I went ahead and created a Google Docs spreadsheet and imported all the sentences from the Basic book: https://docs.google.com/spreadsheet/ccc?...FpHUDBYUEE

I've split the first 1000 sentences into 10 chunks.

How to edit:
- Get your copy of the book or get a backup copy here http://www.mediafire.com/view/?4vy00bnv502kqwc (please buy it if you don't have it)
- Post here and claim a chunk, e.g. thedoc - 1-100
- Check whether the sentence, the translation and the reading match the book. Make a note of the correction on the side.
- Bold the grammar point. (EDIT: the first 195 sentences don't have a grammar point)

Example:

Quote:私は良子に花を上げた。
I gave Yoshiko flowers.
私[わたし] は 良子[よしこ] に 花[はな] を 上げ[あげ] た 。
Basic

From the book:


Correction:

Quote:私は良子に花を<b>あげた</b>。
I gave Yoshiko flowers.
私[わたし] は 良子[よしこ] に 花[はな] を あげた。
Basic

Please post your email here or PM me so I can give you editing privileges.

thedoc - 1-100

Sent you a message through the googledoc.

I'll claim 101-200. Just may take me a bit (finals next week and all).

thedoc - 1-100
nohika - 101-200

Request approved

I'll take 201-300.

It may also take me some time as I also have finals next week.

Chadokoro_K Wrote:I'll take 201-300.

It may also take me some time as I also have finals next week.

Great, request access from the document or send me your email.




thedoc - 1-100
nohika - 101-200
Chadokoro_K - 201-300

Hi!

I'll take 301-400.

@gombost, invite sent.

@nohika, I forgot to mention in the first post that the first 195 sentences don't have a grammar point.

@everyone, would you also like to make a cloze field? Like:

Sentence: 私は良子に花をあげた。
Meaning: I gave Yoshiko flowers.
Reading: 私[わたし] は 良子[よしこ] に 花[はな] を <b>あげた</b>。
Tags: Basic
Cloze: 私は良子に花を[...]。

..to be able to create cards like:

Front:
私は良子に花を[...]。
I gave Yoshiko flowers.

Back:
私は良子に花をあげた。
I gave Yoshiko flowers.

thedoc Wrote:@everyone, would you also like to make a cloze field?

No, we should leave this to the users. Many cloze deleted sentences would become ambiguous this way. Those who need cloze deletion can add a new field to the note with a few clicks and then come up with sensible hints for themselves.

thedoc - 1-100
nohika - 101-200 (?)
Chadokoro_K - 201-300
gombost - 301-400
prink - 401-500

I don't think I will get to it today. I might tomorrow or Wednesday, but I definitely will by Thursday.

prink Wrote:thedoc - 1-100
nohika - 101-200
Chadokoro_K - 201-300
gombost - 301-400
prink - 401-500

I don't think I will get to it today. I might tomorrow or Wednesday, but I definitely will by Thursday.

I probably won't be able to tackle my part until this weekend, so. Too much due before then!

Great initiative thedoc, congrats. Although you do have my moral support, I don't think I would be able to help you guys – I don't have to much time to spare

On the cloze deletions issue, I agree with gombost, both Anki (1.x, 2) have an easy way to create the cloze deletions, so you shouldn't need to create a separate entry in your spreadsheet but you will need to create a separate field in your deck – just an empty one, which has to be right next (after) to the field with the original sentence.

Also, gombost is right saying that people most likely would like to create their clozes in different way. Below, I provided you with couple of examples from my version of DoJG deck.

Also, you could consider tagging the entries. Again, in my deck, I use tags like:
KeySentence, Example, Note, ばかり, 出す, たい, etc.

The examples:

Tags: KeySentence あげる（１）
Q: 私は良子に花[...]。
I gave Yoshiko flowers.
A: 私は良子に花をあげた。
私[わたし] は 良子[よしこ] に 花[はな] を あげた 。
---

Tags: KeySentence が（１）
Q: [...]。
It's raining.
A: 雨が降っている。
雨[あめ] が 降って[ふって] いる 。
---

Tags: Example ばかり
Q: [...]友達が手伝ってくれました。
About ten friends helped me.
A: 十人ばかりの友達が手伝ってくれました。
十人[じゅうにん] ばかりの 友達[ともだち] が 手伝って[てつだって] くれました 。

I'll take 501-600.

In the main grammar section, it seems like nearly every key sentence is presented in both plain and polite forms, with a slash between the two. Do you have a preference of one over the other, or should we just stick with whichever one was included in the current deck?

Another suggestion I want to add is possibly including the page numbers each sentence appears on. That'd make it easier to look up grammar explanations. The tags of the original deck indicated which book they were in, so a field for page numbers is all that would need to be added. It's more work though.

nohika Wrote:

prink Wrote:thedoc - 1-100
nohika - 101-200
Chadokoro_K - 201-300
gombost - 301-400
prink - 401-500

I don't think I will get to it today. I might tomorrow or Wednesday, but I definitely will by Thursday.

I probably won't be able to tackle my part until this weekend, so. Too much due before then!

Sounds great! I don't know if you saw, but thedoc pointed out that the first 195 cards don't have grammar points. They're the sentences from the first book's introduction, so they don't fall within any of the main entries. I think they're perhaps still doable if you read the introduction alongside them and pull out which grammar they refer to specifically.

prink Wrote:Another suggestion I want to add is possibly including the page numbers each sentence appears on. That'd make it easier to look up grammar explanations. The tags of the original deck indicated which book they were in, so a field for page numbers is all that would need to be added. It's more work though.

I think this is a great idea.

thedoc and others, what do you think about including page numbers?

Chadokoro_K Wrote:

prink Wrote:Another suggestion I want to add is possibly including the page numbers each sentence appears on. That'd make it easier to look up grammar explanations. The tags of the original deck indicated which book they were in, so a field for page numbers is all that would need to be added. It's more work though.

I think this is a great idea.

thedoc and others, what do you think about including page numbers?

I thought about the same, looking up by starting letter is annoying, page numbers make it so much easier. +1 from me

EDIT:
I added page numbers to the first 100 sentences. It only took a few minutes since every page has X sentences. It's worth the (negligible) extra time IMO.

Column D is for page numbers.

If you claim a spot remember to post or send me your email address to grant you editing privileges.

001-100 done


thedoc - 601-700

i would love to be involved in this! If there is any job I can do then tell me.

I know it was me who called out these mistakes but I still think the person who made the deck did a great job and that it is a great tool for learning Japanese grammar.

I'd love to help out too!

Schokomilch - 701-800

ill do 801-900

thedoc Wrote:The problem is that it contains typing mistakes

- Check whether the sentence, the translation and the reading match the book.

finding mistakes by just comparing to the book can be difficult

another way to do it is to enter everything again and perform an automatic comparison with the current version to find all the differences, and then manually resolve all the discrepancies by referring to the book

While just about every sentence needs to be touched up, the errors fall into predictable patterns and are very easy to spot after you do a few. It's usually an incorrect use of furigana(食べる[たべる]　vs 食[た]べる), the use of kanji in the deck where furigana was written in the book(友達が欲しい vs 友達がほしい), or the use of 僕 where 私 was written in the book. It's not too hard to do a find-and-replace to take care of most of the mistakes, and a careful reading catches the rest. Even though there are errors, the rest of the transcription outside of those three patterns is generally of very high quality. We definitely don't need to start from scratch.

Jaxon Wrote:While just about every sentence needs to be touched up, the errors fall into predictable patterns and are very easy to spot after you do a few. It's usually an incorrect use of furigana(食べる[たべる]　vs 食[た]べる), the use of kanji in the deck where furigana was written in the book(友達が欲しい vs 友達がほしい), or the use of 僕 where 私 was written in the book. It's not too hard to do a find-and-replace to take care of most of the mistakes, and a careful reading catches the rest. Even though there are errors, the rest of the transcription outside of those three patterns is generally of very high quality. We definitely don't need to start from scratch.

^ This.

I've noticed there are many unnecessary spaces in the reading field. Do you think they can be removed with a search and replace without messing up the furigana?

Example (notice the space before every particle):
先生[せんせい] はご 自分[じぶん] の 家[いえ] で 私[わたし] に 会って[あって] 下さ[くださ] った 。
Should be:
先生[せんせい]はご 自分[じぶん]の 家[いえ]で 私[わたし]に 会[あ]って 下[くだ]さった。

(My understanding is that spaces are only needed to make clear which word the furigana is for. Correct?)

It would certainly be easy to remove the spaces automatically if that's what you want. It's clear to me at least that whatever is in the brackets is the reading of the single kanji or group of kanji immediately preceding it. I've been removing the spaces(my preference), but I haven't used the deck before and I really don't know to what purpose the furigana will be used. Will Anki or an Anki plugin render the furigana above the kanji if it follows a certain syntax, or will it just remain as it's currently written on the spreadsheet?

Jaxon Wrote:It would certainly be easy to remove the spaces automatically if that's what you want. It's clear to me at least that whatever is in the brackets is the reading of the single kanji or group of kanji immediately preceding it. I've been removing the spaces(my preference), but I haven't used the deck before and I really don't know to what purpose the furigana will be used. Will Anki or an Anki plugin render the furigana above the kanji if it follows a certain syntax, or will it just remain as it's currently written on the spreadsheet?

Let me show you.

The original sentence looks like this:

Furigana is rendered (resulting in the removal of the space before the kanji with furigana), but all the other unnecessary spaces remain, like the one I highlighted.

This is the second sentence I posted two posts up with unnecessary spaces removed:

No spaces anywhere.

But we can't remove ALL the spaces because this is what happens:

Furigana gets misplaced.