Wanikani is just not useful enough, but could be


#408

I imagine you talked it through, and I understand that it’s tricky. But if you had even a very simple “Try these items before looking at the lessons” that gave passed items an instant SRS level 6, it could save over 1000 useless prompts each level. That really adds up for advanced learners, so any improvement would make a big difference.

Note that starting items with a higher SRS level makes sure you learn them in a WK fashion, it’s just more efficient for the items that you already know by heart. Thanks for the reply and considering it, anyway.


#409

No one is going pass things like boob grave or raptor cage, or the other WK-specific radicals, so it’s not much use testing those. The radicals would be the issue with letting people skip ahead. So what you just pick and choose what to test people on? It’s messy.


#410

Not only that, but what is a passing grade? Among other considerations


#411

even if you know all the kanji up to level 20, you will still get your 2000 kanji after a year with the crabigator’s help. so what’s the point in complaining?


#412

It’s messy, but is it worse than literal thousands of retests of things you have known for years? I’d rather have an imperfect system that is 60% more efficient than nothing.

There’s no need for a grade per level, or picking and choosing. Just allow any new item to be introduced via test instead of via lesson, and assign it a higher SRS level if successful.

If you come in knowing 1000 kanji and 2000 words, you need to spend over 30000 prompts to get to that level. That’s literal months spent teaching WK what you already know, instead of studying something new.


#413

I doubt it’s possible for such a system to be perfect, but changes can definitely be made to save users time. I think it’s great that we have so many talented Japanese learners on this site, and I’d enjoy WaniKani even more if there were even more of them. If WaniKani’s userbase could be grown to accomodate people who already know a lot of Japanese kanji, that’d be valuable not just for the company but also for the other users learning on the site.

With that in mind, let’s see what can be made of this idea:

Strawman Proposal: Fast Track Level Up Quiz

  • User signs up for WaniKani and opts to take the Fast Track Level Up Quiz.
  • They are faced with a quiz on ALL of each level’s items until they fail to level up.
  • 90% of kanji must be answered correctly to level up.
  • Items marked correct get set to SRS level ??? and those marked incorrect are treated as new lessons.

Some thoughts:

  • This test would probably be difficult to pass (since kanji names might not be exactly what incoming users expect), but at least it avoids the issue of WaniKani having its own unique radical names. It may be worth lowering the bar to less than 90%.
  • This test could be re-purposed for users who come back to WaniKani after a long delay or those who want to de-level.
  • It’s going to be extremely boring to be tested on hundreds of kanji. I doubt I’d want to take such a test. Attempts to sample radicals/kanji/vocab could easily lead to gaps in the test.

#414

Your points are valid, and you’ve obviously put some thought into possible solutions. What you’re probably not seeing, though, is that if you plan to use the mnemonics when you finally reach the stuff you don’t know, a lot of it’s not going to make sense because the mnemonics refer back to stories and such that you will have missed. That’s what Koichi means by “Wanikani specific knowledge”.

There are other dependency issues, too, but they’re more surmountable with some site changes (some significant, others not so much).


#415

Hm, that doesn’t solve the problem of not knowing what the WK radicals are.

I feel it’s better for people who know a lot of kanjis already to simply power through the radicals (i.e. learn way more of them at once), since kanjis and vocab aren’t going to be a problem.

I made a suggestion about that a few weeks ago (it wasn’t particularly noticed, but at least it’s out there).


#416

What you all are saying is valid, but what I’m saying is we have to choose what to work on. And 99% of our users are probably closer to beginners. Although I think it’s a good idea, we only have so many resources. That’s more the issue, than anything.


Please, please let us add our own lessons once we reach level 60
#417

What’s worse is trying to change one product into something else that it’s not meant to be. So rather trying to shoehorn WK into this, something like KKLC or anything in Anki is a better answer.

If you already know the exact 2000 words you’re going to be shown here as the first 2000 words, you already likely have a vocabulary close to the 10,000 range so this site again isn’t for you.


#418

You could also see it as reviewing words and kanji that you’re supposed to know. And what’s the worst that could happen? spending a half a year on learning the other 1000. If your method was already that good that you think you can do it faster, then go for it. But then you would probably not consider signing up for WK in the first place.

In the extreme case, I would also find it useful to sign up to WK even if I’d knew all the Kanji in here, because it’s also a good method for reviewing things. This is even built in into WK. I also regularly unbury a lot of items just so I don’t forget. Let’s don’t be unrealistic: Even if you burn an item here, you will forget it sooner or later.
But yeah, maybe it’s just me… I also like reading a good book twice.

/edit: eating a good book? no, reading…


#419

So then why not force people to learn radicals regardless, but make kanji available to be expedited/exempted. Same for vocab. That would also make the 3 day level up times available for all levels, though. I AM for that, but seeing as that would mean less money for tofugu (in some cases) and a very easy way to abuse the system (which would be the users loss anyway), I wouldn’t expect it to happen.

My thought is simply having a page appear before the current lessons one that essentially looks exactly like reviews. It will contain all the vocab/kanji in your next lesson and gives you can option to claim you dont know it, or enter in the answer for meaning/reading. If you get meaning/reading correct, it becomes guru1, otherwise it continues like normal. Pretty much what felix said, without radicals.

This shouldn’t cause any issues to my knowledge.

Also god damn it koichi lol. Som?!

Also @koichi I just noticed what you said about 99% of people being beginners. Yeah, I honestly can’t argue that and it even applies to me, but an approach like that will be helpful to mid level users (like myself) too. Once people hit the 30s, they will (hopefully) be reading more and picking up vocab/kanji more frequently outside of wk. For a user that reads a decent amount starting at level 25 or so, I would still expect them to be learning hundreds of items that would be later taught on wk. I speak from personal experience on this. Anyone who reads a LITTLE, will know 俺(vocab)、俺ら、俺たち、and 俺(kanji) and thats 4 items right there. 俺 was one of the more obvious ones, but you get my point.

Now, if you considered all of that and still don’t think it to be worth it, I get it. From your reply though, I thought maybe you were forgetting that people can still come across 100s if not 1000 items on here that they knew (at the time of unlock) despite starting as a beginner. Would just save us a bit of tedious work (at your expense) :stuck_out_tongue:


#420

I don’t know. For level 5 and above, going through everything has been useful reviewing.
It’s not any different from someone at a high level (e.g. 60) randomly deciding to reset back to 1 (or 6, as a completely random example).

What I am saying is that you can’t really decide this website isn’t for that category. (But of course, they aren’t supposed to complain about it, the website/Guide/FAQ has been upfront about what would happen)


#421

(Your suggestion is a bit complicated to digest, so I’ll take a look again later.)

If the user fails the radicals, they’d still be treated as new lessons. WaniKani’s mnemonics are intended to fade over time, so it’s not too concerning if the kanji they succeeded on don’t use the same mnemonics. Ideally, they’d also study the WaniKani radicals before tackling new kanji. Since all of the relevant radicals would be unlocked at once, at least a user could (theoretically) speed through them all in 3.5 days, rather than having to do a week-by-week slog.

EDIT: I wish Tofugu had more resources. :slight_smile: Prioritizing for new kanji learners makes sense since WaniKani is geared towards them. It’s difficult to judge what the benefit there would be if resources were spent to improve the UX for those with significant kanji knowledge.


#422

I… don’t understand. Is eating not how it’s done?


#423

The website isn’t intended for that category, doesn’t mean people can’t use it for that reason.


#424

I see that, and it’s a fair enough reason. Just saying that a simple solution might solve most of this problem and allow you to make money from people like me.

I think your site has all the functions needed to implement something like what I described very quickly, so maybe you’ll consider it at some point. I’d be down to beta test it, by the way.


#425

Ah, sorry. I understood “isn’t for you” as “you shouldn’t be here” instead of “you were not the target they had in mind”.


#426

Anyone can use it for that as long as they realize they’re not the target audience.


#427

30,000 Kanji is a lot, but say aye if doing WK has become something of a hobby in that time. (Which happens to be a useful side effect for SRS)