I have a collection of responses I’ve been making since a few posts ago
I like the additional features you’ve put in manga kotoba, I need to go play with that!
And your handshake / applause trick is just in time. I only noticed last week how similar those are
I want that little knowledge bean
八つ当たり was definitely in that camp for me - never noticed it before I learned it, saw it at least now and then after
what is known discussion
This is tough, I agree. So far I’ve decided “what’s known” is a) multi-tired and b) an evolving scale that changes as I study, depending on what use I’m getting out of that information
Right now, I just like knowing whether I’ve learned the kanji for the words I already know and can distinguish them from similar kanji.
Later, I’ll probably care about gathering up more readings, etc, but that feels more like a secondary and slower process of srs mining as I come across the words. At least at that point I recognise the kanji and have a higher chance of guessing the meaning or at the least searching for it quicker (paper reader).
The kanji study app has three “levels of known”, maybe that would work in manga kotoba? If there are multiple levels, the user can define for themselves what that level is, or you could as automate then which could be interesting, ie, level 1 is one reading known, 2 is at least 1 on and kun known, 3 is all words in target vocab known, or something like that.
For me my basis level 1 (is the kanji known) means:
I’ve learned a mnemonic and can distinguish that kanji from other similar kanji
I’ve practiced writing it a bit (helps with above but I don’t drill production).
I know at least one word with this kanji in it (and recently refreshed/added to Anki so I get targeted SRS reading practice)
I don’t spend a lot of time upgrading kanji to level 2 or 3 but I have a vague idea about using that in the future to filter for better levels of ease and knowing more readings
What finally cemented that for me is reading 人狼サバイバル - there’s a character called 伯爵 who gets mentioned often - not actually the same kanji, but clearly some shared phonetic component.
There’s actually a decent number of words and kanji that have got cleared up by disproportionate use in one piece of media for me - 本好き gave me 門 and 本棚、Frieren gave me 怠ける, etc.
I’m feeling the desire to add my mnemonics too… 拍手 is the thing you do at a show, 握手 is the thing you don’t want to do if you just achoo’d into your hands.
I’m finding more and more instances where I encounter a word I don’t recognize in the Maruko kanji book, such as 口実, only to encounter it in manga I’m reading soon after.
The Conan book is easy to make progress on as it’s six kanji on one page. But also, that one feels more like an exercise in looking up words I don’t know so I can complete sentences, rather than helping me recognize kanji. And it can be a little time-consuming.
The Maruko book is easy to get through, but between a three-page manga to introduce the kanji, and typically three kanji per chapter getting one page each, it’s typically 12 pages to get through six kanji.
To reach my goal today, entirely Maruko, I need to get through two kanji pages, the next three-page manga, then one more kanji page. (But I want to get a little ahead since it’s the weekend.)
Reading
While I aim to read less in 2025 than in 2024, I do want to catch up on book clubs and stay on track. For this, I took my current method of tracking my yearly goal and added tracking per-volume deadlines. This lets me see how many pages per day I need to read to finish a volume with the club.
Not including Blue Box (where I’m accidentally a little ahead, so I’m pausing that one for a bit), my book club volumes are sitting at:
The Maruko book sometimes feels like it takes a while to make progress in.
Even though it’s one page per kanji, each bundle of kanji has a three-page comic at the start. If I need to read two kanji pages to meet my progress goal for the day, there may be a three-page comic between those two, making it five pages to go through.
Another aspect that makes it take longer is that I’m trying to add words for each kanji reading to my flash cards. I may skip a reading if it doesn’t seem to be common or used in words I’m likely to encounter, but so far (counting flashcards I created the last few months of 2024) I’m up to over 120 flashcards. (I know some people create that many every few days, but I’ve never been good at absorbing a lot of new words.)
For the Conan book, it continues to be an exercise in encountering words I don’t know and trying to complete sentences that I can’t fully understand while there’s a missing word.
But more importantly, many of the sentences use an expression or idiom that requires the missing word to complete. It’d be like presenting the sentence, “Tim let the ___ out of the bag about my surprise birthday party,” to someone learning English and telling them the missing word is one of: tractor, bridge, cat, light, forest, time. It’s impossible to “solve” if you’ve never heard the expression before.
On the reading front, I’m…probably not going to catch up with book clubs I’m behind in any time soon. But I’m still working on steadily moving forward.
Maybe I’ll use my three-day weekend to catch up a little?
I’d also like to start making progress with the Conan manga series by setting monthly goals so I can keep on track reading throughout the month, but where’s the time? And that wouldn’t help me catch up on book clubs…
Back to kanji, over 95% of what I’ve been doing has been review. It’s been taking up time and I’ve been creating flash cards. I feel like when I get to February and have a larger number of gaps to fill in, I likely will slow down drastically and will start drowning in flash card reviews. I’ll know for sure in a few weeks!
And my time management certainly isn’t helped when I go to do kanji studies and I think:
“If I do this to my site’s search feature, I can reduce the number of search queries being sent to and processed by the server. Let’s implement that and test it right now.”
Then as I’m getting into kanji studies:
“When viewing a word on my side, if I add grade levels to the word’s kanji list, I can see if it would be good to create a card for to help with near-term kanji learning.”
I am always in awe of people able to organize their study log posts. I could give it a try, but that would take away an hour of what can otherwise be reading time, so…
organize means various things to various peoples… one persons mess is another persons organization. Always thought your posts were fairly organized…if it’s that annoying though…
Spend 2-3 hours once writing for a script/template and then it’s a 5-10 min effort … not unlike the weekly bookclub post.
(says the one that isn’t a programmer and doesn’t keep a study log)
Nice! I honestly think sticking with it from the beginning like you are is impressive and will pay off. It must be nice to have some early successes like with 八つ当たり
I can’t wait to see how it goes when it’s more like 30-40% new material
You hit the nail on the head. I was trying to think how to describe my experience with it but never really managed to articulate it, but it’s definitely this. Still enjoyable, just not what I was expecting
I’m curious, do they gradually remove furigana for previously learned kanji?
I guess you could relate to that practical limit rather than the # kanji per day?
I love how useful your distraction tasks are, although that’s probably more likely to sneak past your defenses and win
Umm, yes, definitely takes time from reading, don’t get sucked into it! But it is fun and oddly motivating to have a format, so if you do here’s what makes it easier for me, at least in retrospect - although I’m definitely not the prettiest or most organised log.
My glorified copy and paste method
Introduce only one organizational improvement each time (ideally copied from someone else).
Set your latest update as the solution.
When you write the next update, reply to the solution post and use the loop tool (upper left in editor) to copy the whole post and update each section.
That way, I basically never really do much more than run through the prompt list from the previous update.
If a section doesn’t work I delete it.
And I use <!--- hidden text --->
to hide sections I skip, then I don’t waste time formatting them later.
It’s a set template, with some sections that may or may not be used from one page to the next.
I could manually create each page completely by hand, typing out each section of the template, which is what I did in earlier days.
But instead, I decided to make a web page that I can type the values into, it auto-saves as I go, and I click a button to download the input as a text file in the template format:
If I ever decide to use a template for study log posts, I’ll probably start by copying that text entry page, then extracting the code into a separate file that can be used for both my site’s page and the study log template, and so on.
But for now, it’s back to reading. I just worked through three chapters of a Detective Conan storyline. Should be one or two more left to finish the case.
As per recent posts, I’m playing around with developing a more consistent structure. It’d be nice if the organization of the content of my posts were consistent in something other than their lack of consistency.
No template yet, but I am in the early stages of developing one.
Since this page shares code with another page of mine, I took this opportunity to improve the code. Part of that included replacing about 50 lines of fragile (easily breakable) code with five lines and improving the HTML structure.
Kanji
👧🏻 Maruko
I reached the page for 子 and was glad to see a definition for a usage I don’t grasp very well:
Defition: ことばの したに つけだす。
Vocabulary: ちょう(調)子・よう(様)子
I feel like this…tells me nothing.
I know I can’t completely trust an AI LLM, but I figured I’d check in with ChatGPT 4o to see if it might get me at least within some distance of the meaning here.
Not quite what I was going for. I gave 4o a nudge in the right direction, and it gave me:
子 is used to form nouns that denote a state, condition, or manner. Its original meaning of “child” is largely grammaticalized and abstract, helping to create words that describe qualities or situations.
True? False?
I asked if the original sentence from the Maruko book applied to this usage:
I imagine 「下に」 makes more sense when the Japanese word ending in 子 is written vertically.
I asked for 4o more examples of this usage of 子.
First result:
I have questions.
There may be such a thing as taking abstraction a little too far.
Or else I know nothing about Japanese.
At any rate, Wiktionary does seem to agree with the general “meaning” portion:
I don’t mind as much, but I do want to finish the grade one material by the end of January. It’s six more pages (with six kanji per page), so if I keep up at least one page every other day, it’s not an issue.
Afterwards, I’ll set this one aside as it’s not contributing any to kanji learning. I may continue it at some point in the future, but not in 2025.
SRS
🗃️ SRS
Why is there no flash card emoji? Clearly the tech companies are conspiring against spaced-repetition learning.
(I just noticed a missing r. That’ll be fixed soon. Glad I’m the only one who sees this.)
I think this is the first day I’ve had as many as 32 reviews pending since I added this (admin only) feature to my site.
Some new words gave me trouble as they use uncommon readings for kanji, leading to a five-minute review session. If those cards become leeches, I’ll go ahead and delete them.
…not that I have anything implemented to indicate a card is a leech, nor any functionality to delete a card.
Reading
📚 Current Reads
Plan for Saturday: catch up on book clubs.
Reality: read five chapters of Detective Conan (about 87 pages).
Although I saw the anime adaptation for this case over a decade ago and read the English version of the manga…probably about a decade ago ), I didn’t remember the story/case until about four chapters in. It’s nice to understand the material based on what I was reading rather than remembering.
But I also had to take a nap after four and a half chapters because I was encountering pages with panels like this:
The next case is one I do remember as it’s a favorite of mine, but it’s also a very bittersweet story without the sweet, so I’m not certain I’m going to feel up to reading it in Japanese. But I’ll push through when I get to reading more Conan.
📖 Book Clubs
Since I skipped out on book club reading Saturday, I made sure to get in two chapters of Shadows House Sunday. I figure I’m too far behind on other no-furigana manga book clubs to catch up in them, so I might as well not fall behind on his one!
Also, I completely disavow my namesake character in this series, but that should go without saying since even before this latest volume.
Programming
⌨️ Programming
I haven’t written much about how I write code to streamline my process of getting manga free-preview volumes to create frequency lists from to add to my Manga Kotoba site.
Since Kobo’s search interface is terrible, I use BookWalker to look for freebie volumes.
I wrote a UserScript that de-emphasizes everything that isn’t free (or is a short free preview):
Although, I’m typically viewing only freebie pages where everything is free.
If there’s a series I have at least one volume for on Manga Kotoba, and a volume I don’t yet have on the site shows up as a freebie, it gets a border so it stands out a bit:
I don’t think it’s wrong here, but 子 is also used for marking various increments due to it represent the mouse/rat in the Chinese zodiac, which is also the basis for the old way (12 increments) in old Japan.
KKLC entry #94 characterizes it as a “Noun suffix” without any more fanfare, which sounds like what the above is getting at. Although it didn’t help me understand it, it did help me put it in a bucket, which felt like a more acceptable place to ignore it from than outside that bucket
Whoa I did not realize Conan was so wordy
I like your log structure!! And am continually amazed at your programming refinements
I did learn that in Chinese 子 attaches to a verb to turn it into a noun. It’s possible that its appearance as a suffix in some Japanese words is a holdover from that.
I had asked ChatGPT about 帽子 (hat) but didn’t get much information other than this word’s use of 子 being similar to 椅子 (chair), 箱子 (box), and 靴子 (shoes). It noted the latter two were archaic spellings in Japanese, but I never know whether I can trust ChatGPT or not.
It absolutely requires building up one’s reading stamina to survive.
Then I remind myself that of the 1,860 manga series I have frequency lists for on Manga Kotoba, Maruko is the fifth highest in text density. It’s nearly double that of Detective Conan!
Grade two covers 160 kanji, twice the amount from grade one.
Thus, I expected grade-two to take at least two months. Maybe three due to having a higher number of kanji with meanings/readings, I need to learn better.
However, there are some differences between how kanji are covered in the grade-two book versus the grade-one book one that will impact my pace:
Book 1
Book 2
New Kanji/Comic
2–4
5–8
Sample Sentences
5–8
3
Comics introduce twice as many kanji, and sample sentences are reduced in count, fitting three kanji on a detail page. This supports a faster pace so long as my flash card count stays manageable. With the grade-two kanji split into 27 comics, the best-case scenario would be completing the book in one month.
I started using Migaku’s Japanese Academy last year:
However, I didn’t stick with it because:
Early material is over 95% stuff I know.
Many unknown words I encountered contained kanji I don’t know.
I knew a kanji course was in development, so waiting for that made it easy for me to slack off from continuing the Japanese Academy course and keeping up with its SRS reviews.
From the kanji course preamble:
This kanji course is designed so that if you follow it alongside the Japanese Academy Course, you’ll always learn the kanji for a word before encountering a word that uses these kanji in the Academy Course.
I dropped the Conan kanji book in the last week of January, replacing it with Migaku’s kanji course.
Initial impressions:
WaniKani’s mnemonics are better.
I’m trying to utilize mnemonics, but I’m already struggling with them like I did with WaniKani.
No specific plans for Migaku other than to do reviews and occasionally lessons.
Reading
📚 Current Reads
In January, I read 7.4 volumes worth. That’s across 15 manga volumes, three of which I completed:
At my current pace, I’m on track to complete 84 volumes in 2025.
📖 Book Clubs
I made sure to get in two chapters of Shadows House Sunday. I figure I’m too far behind on other no-furigana manga book clubs to catch up in them, so I might as well not fall behind on his one!
To make it a little easier to view the Maruko book’s kanji information, including the mini kanji section for grade-two kanji from the first book, and have easy access to the kanji’s page on Manga Kotoba, I put together a simple web page:
When populating the kanji list, I found I recognized most of the grade 2 kanji and at least one reading for each. Good sign going in, and I look forward to filling in any gaps.
I also put together a UsreScript to add Manga Kotoba links to Ichi.moe results:
It’s common for me to look up words on ichi.moe if I’m at my computer, and this will make it that much easier to see if a word show up in manga I read, and to create a flash card from it.
I missed kanji learning for a day (only did the comic) due to having to walk to the post office to pick up a package, and then that turned into multiple days of keeping busy with other things and not getting to kanji learning… No time for it tonight, but I’ll need to pick up the slack a little over the weekend if I can!
I’m also considering getting back to doing the kanji learning during my breaks at work. Sure, I won’t be making as much use of that nifty web page I spent a lot of time building up, but it’s more important to keep making forward momentum.
🌏 Migaku
I’ve barely gotten into Migaku’s kanji course and came across 憂 as a kanji lesson.
Since I’m not familiar with this one (WaniKani puts it at level 55), I thought it would be useful to screenshot every instance of the kanji in every manga volume I’ve ever read.
Considering I’ve read over 450 volumes of manga in Japanese, I expect to be taking screenshots for a few days:
It’s reaaaally early into the course for me to be facing the same problem I did in the mid-teens in WaniKani: I’m encountering kanji that basically never shows up in anything I’ve read, am reading, or plan to read…
Migaku is a bit backed into a corner here because:
The kanji course was created to teach kanji covered in their grammar(+vocabulary) course.
The grammar page on 〜になる has the sentence 「僕は優しい刑事になる。」 which is fairly early on in the grammar course.
Thus they have to teach 優 fairly early on in the kanji course.
Migaku breaks 優 down into 亻 and 憂. (WaniKani builds it up from イ, 百, 冖, 心, and 夂.)
Thus, Migaku must teach 憂 fairly early on.
It doesn’t help much that Migaku has 憂 as grieve, and one of its components is 㥑 as melancholy, two English words that, while different, are similar enough that I won’t be able to keep from mixing them up.
Reading
📚 Current Reads
Me reading on the bus this morning:
“This chapter is foreshadowing an onsen scene in a future chapter. But it won’t be until the male lead goes on a business trip a bit later on, so I’m safe for a while.”
I reached the top of the last page of the chapter just before it was time to get off the bus.
Fast-forward to my lunch break at work, and I finished up that last page.
“It’s nowhere near time for that business trip yet, so I’m good to read” swipes page* " the next chap—" quickly swipes back to the previous page.
Why is the cover page for the next chapter the female lead and the male lead’s female co-worker in an onsen? The female lead doesn’t work at their office, so she won’t even be on the business trip. There’s no reason to have the two female characters at the onsen, especially when I’m trying to get some reading in at work.
Needless to say, I did not make any further progress in reading たいようのいえ today.
I opted instead to read 思い、思われ、ふり、ふられ on my lunch break and bus ride home.
As it turns out, the title page with the onsen scene was just a chapter title page, with a safe first page of the chapter after that.
But I should probably read 舞妓さんちのまかないさん Friday to try and get caught up with the club before it’s (eventually) time to start the next volume.
Woops where’s that been hiding! I knew it must be high but
Is it possible to skip kanji in Migaku and revisit later? Just an idea, as KKLC has this same problem, and my way of dealing with it that I came to like most is to read those entries (for kanji that are building blocks but less common), and if I already know a word with it, great, I’ll learn it. If it doesn’t look helpful to me at all, I take what I can from the mnemonics and move on (not putting it in srs or marking it as studied so I can come back to it). Before I started doing that, all those cards for low frequency words I basically never encountered in the wild - leeched out. So if Migaku had leech management maybe it’s not a huge problem, but for me, and I’m guessing you’re similar, I decided to axe the whole category at the source and fwiw I haven’t noticed any downside - on the contrary I can easily study more kanji than I care to srs which has been a game changer for me.
I never added up the number before, but for that post I checked the number of items in my “competed reading” list. I did some rough math to factor out non-manga (children’s books) to get a ballpark total. It’s certainly a higher number than I was expecting!
Breakdown
2018: 1 volume
2019: 30 volumes
2020: 45 volumes
2021: 62 volumes
2022: 91 volumes
2023: 95 volumes
2024: 120 volumes
2025: 4 volumes + 5 volumes worth of partially-read volumes
Totals to 448 or 453, depending on how one counts my partial progress across 12 volumes I’m in the middle of.
With some volumes being omnibus (Aria) or re-releases that compile into fewer, bigger volumes (Sailormoon), one could suggest a slightly higher number, if that mattered. (But the only volume count I’m focused on increasing Manga Kotoba’s!)
If I went by the number of series I’ve read and completed, that’s a much more modest 27.
It is, as kanji is a separate course from grammar+vocabulary.
But then I’m getting cards with kanji I don’t recognize.
I recognize 優しい when I see it if there’s furigana. And I may recognize it without furigana based on context. When I see it all on its own, I don’t recognize it at all.
I may end up pausing Migaku and giving extra focus on the Makuro book. But I’ll make that decision a little later down the line.
I have considered this, using Migaku to look up the cards for kanji I already know, and ones I encounter in the Maruko book, and read up their mnemonics, etc.
I just wish Migaku had a way for me to “study” cards and add them to SRS out of order.
I wasn’t totally sure what you meant by this, so then I wasn’t sure if you knew which one I meant skipping. What I meant was learn 優 since it’s in lots of common words like 優しい, 優れる and 俳優, but skip 憂 even though it’s presented as a building block. I felt like it was still useful getting the mnemonics for it and then connecting that to the next kanji, but there’s no need to srs the less common building block to benefit from the mnemonics if that makes sense.