Beginner Japanese Book Club // Now Reading: シャドーハウス

In the English book club I’m a member of, we’re extremely skilled at getting ties. On more than one occasion, even the second-round vote has ended in a tie, so we’ve generally resorted to picking names out of a hat. Or M&Ms out of a bag, since that’s usually what we’ve had on hand.


I will try to participate this time for real.


I voted for Fruits Basket first so I’d have time to learn a few more kanji for Laid-Back Camp and its lack of furigana :sweat_smile:

(I should be around level 20-something in 12 weeks, so probably still not enough but probably less painful than level 15)


We’re only gonna read the first volume of each book tho, right?

That is correct. Often volume 2+ will get an offshoot club for anyone who wants to continue reading.


Okay thanks!
I’ve seen some posts of the other bookclubs so I figured something like this.


Looking at the first five pages (which is definitely not representative of the whole volume, let alone the first chapter), at level 15 you should recognize 77% of the unique kanji, and 82% of the overall kanji (including repeats). This is not counting kanji in names.

By level 20, you should recognize (in those five pages) 82% of the unique kanji, and 86% of the overall kanji.

Going right to left, you should be able to read everything in black by level 15:


Also worth mentioning that 薪 and 逞 aren’t even on WaniKani. Plus 伊藤 is a name.

P.S. To write this post I searched jisho for “firewood” because I didn’t know the reading for 薪 and I typed 伊藤 by typing , deleting the 勢, and then typing ふじ since I didn’t know how to read that name. (Looks like it’s probably いとう.)


It is. :slight_smile: Or at least the ones I met were called いとう.


Jisho‘s name search claims it could also be いと or いとうざき (whut?) or something, but all the real-world examples seem to use いとう, so that sounds good enough for me.


Ohh, nice work. Wonder if there’s an easy way to get a list of all the kanji in the volume…

The presence of 藤 as the second kanji in a family name is pretty much the only sure sign I’ve spotted that the name should be read in on’yomi. 佐藤, 伊藤, 加藤, 斎藤1 and 後藤 are all in the top fifty most common Japanese family names, and they’re the only ones in the top hundred that are entirely in on’yomi.

1Also written as 斉藤 - and pay attention to that one, we’ll see it later.

Computery stuff

There are various tools out there that each alone brings us part of the way, but they’re different applications on different platforms.

You can access the first two chapters of ゆるキャン△ digital on this 立ち読み. I manually cropped them to the text alone (one word balloon per image) to remove the images, although some OCR software can probably do this. (I haven’t gotten any such software to work on my Linux computer.) I then ran the images through tesseract with the following command:

tesseract input_file.jpeg output_file -l jpn --oem 0 --psm 5

Out of the dialogue in my screenshot in my prior comment, I had to correct I think three kanji, out of 53. (And various kana, numbers, and punctuation.) I could train tesseract to recognize the fonts better.

There’s an Android app called OCR Manga Reader that lets you load an image, tap on dialogue, and it’ll overlay a user-adjustable rectangle over the text (intelligently recognizing the whitespace borders around it). From there I could copy the text (which on my setup shares a clipboard between my smartphone and desktop computer) and paste it into a text file on my computer. (I haven’t tried this, but in theory it should work.) OCR Manga Reader uses a dictionary, so it should give better results than Tesseract. (I don’t know if Tesseract can use a dictionary, but it probably can.)

There’s also a program written in Python that’s supposed to recognize the boundaries dialogue in manga using OpenCV, OCR it using Tesseract, and run it through Google Translate. I can’t get that working on my computer to test the first step.

If there was a way I could extract the text automatically without having to worry about incorrectly read kanji, I would definitely buy the volume and run it through even though I may not read it any time soon.


Missed the main vote, but chose Yuru Camp over Furuba in the mini vote simply because I’ve never experienced it. I’ve love Furuba to death since the early 2000’s, but since the new anime came out just a bit ago, I’d rather wait a bit before reading it.


Thank you so much for this analysis! :blush:

Even if it’s only of the first five pages, it does give a glimmer of hope that I might manage with some help. It looks like leveling up to 20-something probably won’t change things that much though, so reading this pick first or second won’t make much difference…

I guess I could always buy it and start with the club, and put it aside for later in the year if it gets too painful at my current WK level!


I’m not interested in Laid-Back Camp but I still don’t want the club to read Furuba first because it’s more likely for me to join Furuba if reading starts closer to the second half of the year.


You are welcome to vote based on whatever personal motivations you may have :grin:

@skymaiden I honestly seem to have lost my spreadsheet of the vocab from the first two chapters of Laid-Back Camp, which is baffling and annoying. However, I do have a rudimentary Kitsun deck (mostly focused on the kanji-containing words) which I could publish if that would be helpful? I might also be able to export it and thereby recreate my spreadsheet…

I confess also that I remember there being one kanji I couldn’t successfully look up :thinking: I’ve been waiting a year and a half to be able to ask someone in the book club for help :grin:


Hehe, that was my motivation for voting in the second round to.

1 Like

Oh noes! Maybe we could make a new spreadsheet like we have for some of the other book clubs?

The collaborative decks are a killer feature of Kitsun, but unfortunately for me it’s horrendously expensive on top of all my other Japanese-learning resources (lifetime costs as much as I paid for WK, and I’d rather buy books now), so I won’t be able to use that…

Oh, we definitely will. It’s just that this would have been useful to look at in advance.

1 Like

I would loooove it if you could try and export it so we can see the vocab you picked out from these chapters! :pray: