I’m really failing here at making myself clear, I guess… this isn’t something I’d make a feature request of, any more than i would make the feature request “if I type foundation, but I’m thinking of the makeup, but the word means ‘an NGO with an endowment’, mark it wrong, but if I’m thinking ‘like the Bill & Melinda Gates Foundation’, mark it right, please”.
WaniKani can’t know what’s in my head. That’s why I need the ability to mark it wrong, even when its simple heuristic thinks it’s right. And I used to have that. Now I don’t. Is that really so unclear?
Fair enough. But I don’t think you understand my point, either. I think your usage of “need” above is overstated. You might learn that one item faster with the feature, but even that isn’t a certainty. It may stick in your head four months hence precisely because it was incorrectly marked as “correct”.
Some things annoy me, what is the difference between a discussion and a conversation exactly
Personal wealth/own funds/personal funds/own wealth caused me grief today, had to add synonyms.
Some things should not be allowed to be synonyms though like the transitive/intransitive stuff (or it should give you a wobble and “that’s not quite right”?)
Are you thinking of 写す? That word can mean to copy something (模写する) or specifically to take a photograph (写真を撮る), but it’s 多動し (a transitive verb) either way. English has different words for the two concepts, but Japanese doesn’t (there are MANY examples that go the other way).
Transitivity is a difficult topic because it works so differently between the Japanese and English.
I have consistent issues recalling which of verb pairs—especially mediopassive pairs—is which. I’ve used ChatGPT to try to come up with mnemonics, I’ve put them in a chart and run regexes, but there are just too many exceptions to almost any rule you might come up with, and at this point I have at least 200 of them active. I do better than chance—but SRS is specifically designed for things like this, where you need to remember arbitrary info that you can’t easily unify with principles and theories. It’s ridiculous that I have to go to some other SRS app to get that when a) the word list is already here and b) I could do it last night!!!
It’s not some crazy out-there thing. I could do it last night with a keystroke. I can’t do it now. My inability to intuitively recall which verb is which did not magically disappear overnight, but my ability to methodically work on memorizing them did.
Stop telling me that I’m exaggerating my own need. We’re talking about my own mind. I know what I need in my mind. I can’t even begin to understand the presumption of someone saying they know better than I do what my mind needs.
Are you a psychiatrist or neuroscientist? Even so, shouldn’t you at least examine me if you’re going to pass judgment on what my brain needs? (I’m a cognitive linguist by schooling, at least…)
To the several people mentioning that those of us unsatisfied with the update are “demanding that WK never updates and should stagnate”, that’s patently untrue. Updates are necessary for any software to improve. My point was that, for a web service that is designed with heavy integration of third-party userscripts (heck, that’s half of what this update was about was making it easier for devs to make the scripts), I think it was a very bad move for them to have the update be mandatory, for everyone, with no option to roll it back until things caught up. Many, many services allow the user to roll back to previous updates in case a new one breaks things on their current setup. e.g. things like Windows Updates and many game titles on Steam likewise can be rolled back, because every hardware/software setup is unique and updates are frequently bugged in relation to a large portion of userbases. Heck, even notoriously user-unfriendly services like Facebook and Reddit have in the past allowed the temporary rolling back of major UI changes.
It’d be different if WK was more like something like Duolingo which does not have TP API support or even customization in general, so that updates - while they might take some getting used to - at least don’t break a bunch of user settings. But given how WK specifically caters to the individual user experience by adding features like API tokens and such… while I wholeheartedly support their endeavor to make updates to improve things for everyone, the actual implementation shouldn’t’ve been a brute-force overhaul with no option for users to keep things the same, at least temporarily, so that things don’t get broken. Call us “dramatic” or “entitled” all you want, but this was a huge detrimental impact to a service that most of us use daily, depend on for learning, and have paid a lot of money to use, and we are 100% justified in being dissatisfied with the abrupt change.
The rest I get but this just isn’t the same. Excessive is ‘too much of something’, unnecessary is ‘not needed’. Something excessive can sometimes also be described as unnecessary, and vice versa, but the connotations and nuance are completely different. They aren’t synonyms at all.
“The punishment was excessive” - The punishment was too strong.
“The punishment was unnecessary” - The punishment was not needed.
I get that people love their synonyms, but I feel like people adding too many hinders their progress. In the same way, I know people who will ignore kana spelling errors, for example between どう and ど, for words like 速度 「そくど」, when 側道「そくどう」is a completely different word that will be misunderstood by natives. People are so obsessed with speedrunning it that they hurt themselves.
There’s a quote about carpeting the planet because someone is uncomfortable wearing slippers.
Apologies, but it sounds to me that you worked hard to make the service work to your liking with ChatGPT, regular expressions, scripts and I don’t know what else, and are now upset that all those things no longer work the same? That does sound sorta crazy out-there to me.
I’m not exaggerating your own need, I’m stating what worked for me and at least a few others.
To be fair, in English “totally” does exactly the same thing as 全然—it intensifies positive verbs in the positive direction and negative verbs in the negative direction. It only seems weird when you try to make it a word-by-word substitute.
Three suggestions for you (getting a bit off topic but this thread is a tad dramatic so offering something to mollify that).
Make a dedicated anki deck for verb pairs that you have problems with. WK can’t and doesn’t do everything. It is for learning kanji.
Consume more Japanese and these pairs will not bother you as much. If your goal is just consuming stuff then this is more than enough.
If you are speaking Japanese frequently then just make a mental note of which verbs confuse you when speaking and look up its transitivity later (or ask the person you are speaking to). Next time you use it you will probably remember. Generally I only need to use something a few times for it to stick.
Huh? Outside of WaniKani, which is an SRS app for learning arbitrary info through memorization, I have tried to find ways—including with machine learning and algorithms—to bring any principles or patterns to bear on classifying the verb pairs, like people do with, for example, rendaku or counters—which are also memorizable but you can cut down on the memorization required if you learn a few principles. Aside from “す is usually transitive” and “if there’s an え-row okurigana form and an あ-row okurigana form, the え is 80:20 to be transitive”, I struck out finding good rules.
I didn’t expect WaniKani to “support” that, any more than I expect WaniKani to “support” or not support the dictionary I choose to use. It’s an outside thing. You can’t learn Japanese exclusively on WaniKani, it’s just a vocabulary system—I’d wager that 99% of WaniKani users use outside resources.
If you think using outside resources to learn Japanese along with WaniKani is “crazy out-there”… are you including Tofugu? Even so, Tofugu doesn’t carry a full set of grammar lessons, definitely not in a form you could use to learn Japanese.
Or is it because ChatGPT and my regex tools are on a computer, you think people using WaniKani should refrain from other computer-based Japanese tools? That’s an… unusual… opinion, but you’re welcome to it, I guess. I’m going to continue to use outside tools to study Japanese.
Ah, this interests me… Would you suggest in general I should stop doing what others have suggested they find important—to make the distinctions between closely-related words as sharp as possible—inside WaniKani, and only use outside SRS apps for that? Or is that advice specifically about verb pairs? If so, what makes verb pairs different from other vocab people talked about earlier today?
I mean, I could just quit treating 上がる and 上げる and 交じる and 交ぜる like words I need to care about differentiating inside WK and create Anki decks for them—is that what you’d do?
It feels like just pushing the same problem from one place to another except than Anki lets you decide what’s correct vs. incorrect, and WaniKani overnight doesn’t let you do that anymore. Would you have made the same recommendation yesterday, when it was possible to manually decide like you can on Anki?
Sorry for the deluge of questions, but it sounds like you have a very good head on your shoulders for how to go about learning these things and I’d like to hear more!
You are missing the big one “if it ends in aru (ある・まる・かる etc) then it’s the intransitive one in the pair”. With す and aru rule you already covering like 70% of all pairs, I don’t think it’s really that useful trying to find more subpattern.
SRS will help you with that. ERU vs ARU common pattern for transitive vs. intransitive, it’s about our level you need to start paying attention to that.
交ぜる reading comes up again on level 19 also meaning to mix something so that helps.
Yes, of course—I mentioned it earlier in the thread and just forgot to include it there. Good to know that it’s not worth trying to explore further. 30% is just enough to be really itchy and think you should be able to do better. But if you can’t, you can’t. Thank you!
There are plenty of exceptions to this. Believe it or not, I actually side with @TreyE here that rote memorization (with an SRS or otherwise) is the best way to learn transitivity.
Scratch that. The BEST way is to have conversations with Japanese speakers and learn what “sounds right”.
My feeling: rote memorization, mistakes, iteration are the holy trinity. This is the way.
