Hi everyone!
This is a long introduction, there is a TLDR at the end.
I am a long-time learner who started using WK recently. I have read some of the advice on this forum, which have been very useful but I am still a bit uninformed about the pacing, for which I would really appreciate some references.
As a context to this, I have been learning Japanese for about 7 years, I am quite good at most things in the language, however I have often neglected kanji (I took a language exam 3 years ago roughly equivalent to a B2 level, and I got great results only because a kanji dictionary was allowed during the written portion).
Now when I say that I have neglected kanji, I mean that I can still read several hundred in context (this is really difficult to estimate, but Iād say between 400-600), but may mix similar kanji up without context, and only know about 100 kanji well enough to be able to handwrite from memory (I estimate that this will prove to be a huge advantage compared to complete beginners until around level 10-15, after which my workload will increase a lot).
The reason I started WK in the first place, is because I have been accepted for a semester of studying in Japan, and got the information that the entrance Japanese ability measuring test includes a handwritten composition for those that score well on the reading, grammar and listening tests. Fortunately, I have managed to free up a lot of time over the summer, meaning that I have probably 3-4 hours of brain power and time to use on active studying per day, with few breaks when I would have only an hour or two.
I already use a premade Anki deck to structure the kanji I learn by JLPT levels, I handwrite all the kanji, and I choose example words that contain the readings (only leaving out very rare or outdated readings), practice writing them some more, as well as identifying which radicals are involved, and naming those radicals for better retention. I keep track of all this in an excel, and while it takes a lot of effort, it seems to be working quite well. I also regularly attend classes.
I thought that once the WK workload gets high, and I start learning many new kanji I could potentially switch to WK as my main method, along with handwriting practice. In any case, I thought researching this while the workload still leaves room for me to think would be the best.
While this introduction has been way too long, does anyone have advice on planning pacing, their own experiences with taking on a bit too much, advice about how far to think ahead for planning reviews, etc.
Any help would be greatly appreciated! Have a great day, everyone!
TLDR: I started using WK and am already fairly knowledgeable about radicals, kanji and vocabulary. I want to get the most of out the 3 months I have before a test of this knowledge.
I would appreciate any thoughts and references on pacing and preventing review hell, ideally from those with experience on the platform. Thank you to anyone who would kindly share their experience.
Given your circumstances, Iām not sure that WaniKani is the right tool for you. WaniKani makes more sense for complete beginners who are learning Japanese and more specifically Kanji from scratch.
If you can already use the language at an intermediate-advanced level, the very rigid structure of WaniKani will probably prove to be a hindrance more than a feature. Also WaniKani likes to teach vocab using kanji, but in your situation it seems that you probably already know a lot of vocab and just need to associate the kanji to those words.
I think in your case it makes more sense to learn kanji as need arises from the way you use the language, not using premade lists.
What I would do is practice writing these types of composition, see what kanji I need to express myself and then put those on Anki for more targeted practice. I think that would be much more efficient than just going through WK levels or use premade Anki decks (you can still use premade Anki decks for that: just suspend every card and manually unsuspend them when you actually want to study them).
But WaniKani will teach you all sorts of kanji that may be irrelevant for the task at hand. You may not need to know how to write ę¦é and ęÆ and ęøÆ and ē¼é” and ē¼ and ę³£ and ę¢ å¹²ć for your exam.
Hi!
Thanks for your input, it was doubly helpful as it pointed out suspending Anki cards, which I did not know about, and made me realise that I indeed should start with writing compositions (perhaps I was procrastinating on that due to having disliked writing them previously).
However there is one point I forgot to mention. Having been reminded of the importance of kanji, I would like to build up a habit of learning them to make gradual progress towards an advanced level of literacy. For that, I think WK would be great, perhaps given a more leisurely pace, especially until another higher priority goal is in place.
What would you consider a leisurely pace?
Yeah I think absolutely doable, I was mainly taking your relatively short deadline into account and advising more efficient methods of study for this very specific objective.
Beyond that if you like the WaniKani level system then by all means, go for it. It certainly wonāt hurt. Especially the first ~30 or so levels will be full of very useful kanji youāll want to know regardless of what you read.
Keep in mind that WaniKani is focusing heavily on recognition, not output. It doesnāt try to teach you to speak or even write Japanese, but mainly to read Japanese. As such if you want to be able to write comfortably youāll have to continue your drills on the side.
I always found this post bizarre because, when you read what they actually describe, it sounds like a terrible process for an advanced learner. You go through months of having hundreds of reviews per day (since WaniKani doesnāt have an āeaseā system that would let you progress faster through easy content like Anki does) while learning very little.
Even if you manage to go very fast through the reviews because you already know everything (and WK not having undo, you better not go too fast unless you want to risk typos), how fast can you realistically go through hundreds of reviews/day? And remember that if you want to go full speed like they did, you have to do reviews multiple times per day, sometimes at awkward hours.
Reading further, they basically knew almost everything up to level ~50, and from that point onward they learned a lot. Cool, except that those are also widely considered the least useful levels because they teach you more niche kanji that may not appear in the things you routinely read. So you know how to say Red Carp now? Wow congrats!
They reached level 60 with 99% accuracy. Thatās absurd. If you have 99% accuracy in an SRS system, youāre just wasting your time.
I donāt understand how anybody would read this and think āyeah, that sounds like a good use of my time and moneyā. It seems completely inefficient and highly frustrating.
So when I started WK I had about 400 kanji I either completely mastered, or knew really darn well. (And then about 200 more I sort of knew but only if I recognized the vocab, and not consistently. A very low level of āknowledgeā for these.) That puts me around the same level as your starting point.
And personally I have found WK to be a fantastic method. I do use JPDB for extra practice, but I donāt hand write anymore - I used to many many years ago.
That being said I want to focus on the part where you said 3 months. WK is extremely time gated. 3 months was probably about how long it took for me to get to the content I wanted to get to (the things I didnāt know that well), and I was going at a pretty fast pace by selecting lessons manually.
So while I certainly would recommend WK overall, I donāt know if Iād recommend it if your goal is specifically in 3 months.
If your goal, however, is just to review what you can in those 3 months and then keep using WK while for a year or so, then I think itād be a good addition.
PS: I didnāt find it hard to do hundreds, sometimes several hundreds, of reviews per day because the existing knowledge meant high accuracy and fast reviews of just 1-2 seconds. I would, however, recommend you getting an undo and/or getting an Anki style plugin for those first 10 levels (or whatever is appropriate until you start seeing a significant amount of things you donāt know, and then drop the Anki style plugin)
Adding as if you do decide to use WK and blast through the initial easy stages, you might wonder ābut wonāt I get a ton of reviews several months later when theyāre ready to be enlightened/burnedā.
Yes. I started in October and April is when stuff came up for burning, and oh boy there was a lot of it. And I will probably keep seeing them come up until June-July. And here is how I manage that:
First, I change the review order from Random to Descending SRS stage (I use Tsukurame on IOS but there are similar apps/plugins for other platforms)
If I see it has the really easy stuff from the beginning, I do as many reviews until I get to things that are not super easy, wrap up, and change the order back to Random.
Count goes down a lot in just a few minutes and I get a sense for what my real review count is with the easy stuff taken care of.
Same, but I donāt think that makes WaniKani the wrong thing, just that it makes the SRS method the wrong thing. It makes perfect sense to do several levels of WaniKani to get the concept of naming the radicals and then combining those to make little stories to remind you of the kanji. Learning a procedure, not the items themselves.
Then, I would just page through the level pages learning all the pre-made radical names, and random-access read the mnemonic for each kanji I was trying to remember from encountering in the wild. And if itās not on WaniKani, well I do know the radical names and the mnemonic concept.
I donāt think the true value of WaniKani is the SRS to begin with, I think itās the creative set of mnemonics and the building-block learning order. (Even though some of the mnemonics are less good than others)
Interesting take but it must be subjective because for me itās the opposite. I stopped looking at the English mnemonics pretty early on because I found them distracting and counter-productive. IMO they make sense early on to bootstrap the system, but after that I much prefer building my memory devices from other Japanese words and kanji, not very rough English approximations. And I tend to avoid mnemonics unless I really need them anyway.
For me the value of WaniKani is mainly in the structure and āhand holdingā that make it easier to build essential kanji knowledge and give you short term goals. If you already have a good grasp of that stuff I donāt find a lot of value in quirky mnemonics personally.
True for me as well. Itās about building up of smaller radicals into Kanji, assigning names to the radicals/Kanji in order to able to make use of mnemonics, except that you donāt have to use their radical names and associated mnemonics.
Or possibly not to use mnemonic at all. Another goodness (?) of WK is typing strictness. I donāt question on Kana typing as much as English typing. Simplicity makes it easy to mark cards as either right or wrong; and also makes it less painful to go through hundreds of reviews every day.
I would also count separating into levels as a pro, as it is easy to visualize, as well as chunking vocabularies well into clusters. But clustering is questionable, a big list managed well could work too (better SRS management perhaps). And there might be better ways to visualize.
All I want to say that, it can be better, when well managed in Anki. But not managing well could also make a worse system to be stuck in for years.
Itās possible not to replicate with SRS and Anki as well, but that makes it harder to compare pros and cons with WK.
Better SRS systems than the very rudimentary thing WK has allow you to e.g. mark items as already known so you donāt have to review them, or mark them very easy so they get shown to you much less frequently than they otherwise would, which makes it much easier to start with āhere are 2000 kanji itemsā and get to āI donāt see the 500 I know absolutely, I get occasional review of the 500 Iām pretty familiar with, and I get the benefit of the system on the ones that Iām shakier onā. I think WK here is absolutely the wrong thing specifically because its SRS is so rigid and behind state of the art here.
For me, I value being able to understand the algorithm, which is one of the things which drew me to Wanikani. Even though Iām not currently doing Wanikani, Iām doing Goldlisting, which is just as simple an algorithm if not more so
I think you could have āI know this, donāt show it to meā and āthis is easy, push it forward two steps instead of oneā without the algorithm being impossible to understand.
More generally, prioritising comprehensibility is paying a cost in efficiency. In the āusualā WK setup you pay the cost gradually as you go along, which makes it less bad. But in the āI already know half of this and need to catch the system up to the state my brain is already inā situation of this thread you are paying a lot of that efficiency cost up front before you can get to the new items where you get some payoff. So I think it matters more here to prefer efficiency.
I am currently in the middle of setting up a system just like you suggested.
A few days ago I made a document with a list of English prompts for each kanji, and tested myself in how many I can actually write off the top of my head. Since I had several hundred listed, this took some time, but I think it was well worth it.
From this I can separate them into āwill not forget for years at a timeā, āI can usually write this down but I should practice occasionally just in caseā, āI know these on my better days, but I might forget or mix up a radicalā, and āI have almost completely forgot theseā.
Taking this list, I decided to take out the ones that I will not forget for a long time (and often see anyways), and make the rest into an Anki deck where I have the front as the English prompt/meaning, and the back as the kanji (and maybe a few example words). Anki has the very useful function of being able to choose āAgain / Hard / Good / Easyā and the review time implications on top of them. This means that I can intentionally manipulate which ones I want to see more often, and which ones can have a larger review gap. Now letās say on every review I get wrong, I practice handwriting that kanji for a bit, and every week or two, go through the list and see which kanji are still shaky. Then I can focus extra attention on those, which means few kanji will āslip through the cracksā.
I also intend to do this with radicals, although primarily using the names I came up with (potentially including WK names if I find those to be good, or just to strengthen my familiarity with the WK names).
I probably left out some things here but I will only remember those later.
By the way, does anyone happen to know a way to search for kanji or radicals in the WK library?
Or will I have to manually click through Pleasant, Painful, Death, etc. with CTRL + F as my best friend?
