Except you’re not technically wrong in that you’re putting in a correct reading for said kanji, just not the one WaniKani wants. I forget what script it is, but there’s one that will actually notify you. If I were to put in うし for 牛 when it’s looking for onyomi, the box will shake and a popup will say “we’re actually looking for the onyomi here”. Why is that such a hard thing to do? How is that bad? It acknowledges you got it right, doesn’t mark it wrong, and just says, “Hey, do the other one.” If you don’t know it, you don’t know it and still get it wrong. If you do know it, cool. You got it correct and didn’t have something you actually do know marked incorrect. Where is the foul there? How does that hurt the learning experience? If anything, that makes for a better user experience, which is why those kinds of scripts exist. That one, however, should totally be natively implemented.
This is, in my opinion, dumb. I should be focused on recalling the word and getting it right, not worrying about what the background color behind the word is. It should hint me in the right direction, but not punish me for putting in a correct reading for the kanji.
Should WK shake and warn me when I answer with じんじょう instead of にんじょう for 人情? I mean, じん is a reading for 人, and once you tell me that じんじょう is wrong, I’m able to answer with にんじょう. See, I did know it after all.
If you say “well in that case you don’t have the excuse that you didn’t know it was a vocab question, because it had two kanji,” but that same situation can happen with solo kanji too.
How does WK know if you know whether it’s a vocab question or not. Maybe you know it’s a vocab question, but you just don’t remember if ぎゅう or うし is the word for cow. That’s exactly the same as the じんじょう / にんじょう situation I mentioned, but people want WK to warn and shake instead of sending the item back in the SRS when it’s one kanji, whether people realized it’s a vocab question or not.
You’d be able to burn it without ever giving the right answer on the first try.
Sure, when we’re talking about 牛 it sounds so simple that it’s not worth worrying about, obviously people will eventually learn the right word. But not every solo kanji word is that common and simple.
I’ve been here since '15, hardly a new user, and I see no reason to not use override.
No time for the BS. You can produce typos or have a brain fart even if you’re born Japanese and speak nothing else all day long.
All those assumptions by people who do NOT speak Japanese are interesting to read though.
In the end, just do whatever you want 
I disagree. If you don’t remember the reading used in a word, you don’t know the word. This is opposite of knowing readings for a kanji and mistakenly putting it in at the wrong time. IMO this should only happen for single kanji words. The instant another kanji is in use, you should never make that mistake, and if you do, it’s because you didn’t know the word. WK should most certainly not warn you you’re using a wrong reading in a jukugo word.
But @starmech, you are wrong if you tell Wanikani “the word for cow, 牛, is pronuounced ぎゅう”. It is wrong to say that cow in Japanese is ぎゅう.
I really don’t think there is a script like that. Maybe you’re thinking of the Android app? The Android app does this, for kanji questions. Because in kanji questions multiple answers are technically right.
EDIT: you might wanna check out this thread
It’s possible for it to be just that you didn’t notice that it was a vocab question. That’s the situation where people make topics to complain. But it’s also possible that you did notice and just didn’t know which reading is right.
THAT second situation is exactly the same as the one I mentioned.
But the proposed behavior doesn’t distinguish them.
Ah ok. I didn’t think about the possibility of getting the onyomi and kunyomi readings swapped in your head. Makes sense.
Yeah this must be it. I use the Android app as well.
Yep I agree.
It would be one thing if they presented the vocabulary words in the context of a sentence, then reading the on’yomi is 100% wrong. But just seeing the same kanji with a different coloured background is really a pretty abstract criteria.
I understand the rationale behind on’yomi answers to vocabulary being immediately wrong, but it does seem like it’s too punishing - at least for the first few levels. It’s hardly if ever a problem for me anymore, you do get used to treating things with purple backgrounds as if they’re standalone words, but it is pretty arbitrary for the first while.
Mistakes/accidents happen, but they’re just as likely to happen while entering an English meaning as when entering a Japanese reading. That’s a separate issue, and if you want to use the override script for such situations then more power to you.
The issue at hand is just not understanding the question WK is asking. Responding with on’yomi thinking that is what’s expected, and then instantly failing because that wasn’t what they were asking for. And, given the surplus of these posts floating around the design of this system is obviously not 100%.
Arbitrary? Sure. Useful? Hell yeah.
Wanikani needed a way to distinguish between reviewing vocabulary (standalone words) and Japanese characters (kanji). They employed two solutions:
- explicitly telling you which one they’re looking for - kanji or vocabulary (as you mentioned yourself)
- and an additional (arbitrary) visual cue, i.e. distinct colours, that you can get used to over time so that eventually doing reviews becomes almost instinctive
It really doesn’t take that much work to familiarize oneself with the system… I think most of these complaints stem from the misunderstanding of the difference between kanji and vocab (including single-kanji vocab).
I’m sure they could find a better way to present the information. But giving a do-over when you type in the wrong answer for a vocab question is not the way to go. The reality (unfortunately) is that WaniKani has no way of knowing if the user answered wrong because they misunderstood the question or because they actually didn’t know the answer. So to me, it makes sense to err on the side of being strict and always mark you wrong.
i don’t have any issues. every feature that WK doesn’t have that i’d want is available via script. just because i think i’d like it this way, doesn’t mean anyone else would have to put up with it 
Completely agree with you here. If you don’t get the word exactly right, then yeah, it’s wrong. Sometimes I make typos but that’s fine if it marks them wrong. I still think presenting vocabulary in the context of a sentence and doing everything else the same would be a vast improvement.
Of course being able to read vocab in sentences is vastly important, but I think that always having the context of a sentence to provide clues can be another crutch and make it harder to read the word by itself. It’s kind of like how I’ve known the word 綺麗 for ages and have a 100% correctness rate on WK, but I have less than a 90% correctness rate for each kanji, and I still haven’t burned 綺 yet.
That’s a solid argument.
but that’s how it works. japanese people recognize words instantly, while figuring out a kanji takes a moment, if the kanji itself is not a word on it’s own.
you can almost see them rummage through their memories
and then, they remember it not as a single unit, they remember words it pops up in.
the reason is, they don’t learn kanji one by one after the first 2-3 years in school anymore. they learn words only. they’ll never encounter them standalone, so learning them like that wouldn’t make sense.
i think some people don’t know this and just trust that “koichi knows best”, while others are aware and try to be more japanese than the japanese.
learning them as single kanji is useful imo, but don’t beat yourself up for not recognizing them when they pop up with no context on a flashcard website.
I see individual kanji on those little chalkboards they use for teaching the stroke order and important elements (hane, harai, etc) in all elementary school grades, not just 1st to 3rd grade. Obviously they also do drills in booklets that mostly focus on how the kanji appear in words, but all grades do that too.
But once they’re in middle school, yes, they stop doing the thing where the form of new kanji is focused on in a lesson.
some people do this even as adults, when learning for the 漢検. kids also learn them mostly by drill via handwriting, so the situations aren’t the same at all, i get that. what i meant though is that it doesn’t make sense for wk students to lose hair when they don’t recognize a kanji from a word they get right every time.
it will come. sometimes it takes a little, but that’s no problem. that’s where i would recommend the addon “vocab beyond”, at least to people with some vocab knowledge. it shows you a long list of words the kanji pops up in that don’t appear on wk, and if you see the thing in several words, your knowledge of it’s meaning becomes clearer and the memory stronger.
Right, I agree that people need to be able to recognize the vocab instantly even if there’s no sentence surrounding it to provide context, which was what I was arguing for. The kanji was just an example I used because it was a similar situation, I don’t actually care much about recognizing those two super rare kanji in isolation. Besides, WK is already lenient towards kanji readings…