Yeah, anything that moved around in a content update in WK isn’t going to affect an inflight learning in KameSame
That’s a cool feature, thanks! It just needs to be a bit blurrier, I think.
Thank you so much for implenting Vacation Mode!!! I appreciate you so much Justin, and I will do everything I can in my power to contribute to your Patreon. My success in Japanese will be largely attributed in your work, and I will be forever grateful. ありがとうございます！
Hi, I am new to both WaniKani and KameSame and have some questions before I will begin learning:
- Should I first cover WaniKani and only after that KameSame or is it better to do it in parakllel?
- I read that WaniKani has 60 levels and KameSame 100… so how would that work doing them in parallel?
Thank you for your help!
- I’d recommend doing them in parallel. I’d also encourage you to dump any word you come across and want to learn into your review queue by searching for it and learning it or by pasting a list into the text area on the lessons page
- KameSame’s level system is based on XP gained by leveling up items thru the various SRS stages, and it’s not tied to any particular content. Unless you drill an absurd amount of items every day, it’ll take most people about two years to reach level 100
Do your Reviews for both of them
I haven’t started, because I have too many Reviews already
hi, im sorry, i’m stuck on my second ever lesson. KameSame is asking me for the Kanji for “axe,” and i know it’s きん、however, the dropdown list of kanji characters gives me 101 options and not one of them is the actual kanji for axe. is there a way around this?
**this isn’t a complaint about the app, just me trying to figure out the japanese keyboard!
Edit: i just figured out that if i put a number in front of the hiragana it figures it out for me haha hooray for counting bread loafs
This is why KS is that way, so you can figure the keyboard out! Good luck
Major dependency updates today. We went from Ruby on Rails 5 to 6 and from Preact 8 to 10.
You shouldn’t notice anything (but perhaps things will be a little snappier). If you spot any new bugs, let me know!
Because a lot of folks’ item lists are getting longer and longer as they invest more of their time into KameSame, this weekend I wrote a component for item lists that should be able to be spread to most of the parts in the application that list out items (review/lesson summaries, search results, etc). For now it’s just on the current-progress lists from the main page.
- Switch modes (right now, full-width detailed list and compact postage stamp size)
- Filter via text (will expand query to romaji, katakana, and hiragana and match against any of the kanji, reading, and meaning text on each item in the list)
- Sort by item type, english alphabetical, reading phonetic order, WK level, a custom frequency analysis the app has been using for a while
- Stable URLs that update as you use the component so that deep links are preserved
Here’s a demo gif:
Just started trying KameSame today. It looks really nice and feels way better than KaniWani already. However, I keep hearing about an app version and I can’t find it in the App Store, is it available for iOS? Is it still in beta and I have to download it another way? I’d really like to use this on my phone so I can do reviews as soon as they show up, no matter where I am.
Wait, I’m dumb. It looks really pretty in Safari so I guess there’s no need. Thank you!
You can add KameSame to your iPhone’s home screen and it’ll act as a “progressive web application” and adjust to take advantage of the full screen, streamline links to other apps like Japanese & Midori, and so on.
There aren’t any notification features (yet), however.
Wow that was a fast reply, thanks! I didn’t even know that was a thing so I just Googled it and figured it out. ありがとうございます！I look forward to using this more in the future.
Me again! I’ve noticed that the program doesn’t know which reading I enter to input a kanji, so I can enter the kunyomi reading while not recalling the onyomi reading it expects. For example, it asked for 君(くん), but I found the kanji by answering きみ instead. I suppose it doesn’t matter that much as long as I can recall the kanji itself, but I wanna make sure I can recall the correct reading. Dunno if this is addressed in earlier posts but I didn’t wanna read through 800 posts to find one. Thanks!
Yep, the reading you use to produce a kanji doesn’t matter. (Including whether you use a keyboard IME or you draw it with a drawing keyboard.) The test is whether you can produce it or you can’t; I don’t care what readings you know for that purpose.
Hey everyone. I decided to make a dark theme userstyle for KameSame. If you’re interested, you can get it here:
Please let me know how it looks, and if I missed anything. This is my first attempt at writing a userstyle.
Apologies in advance if this was answered before, new KameSame user here.
I’ve entered 入口 as an answer for Entrance (noun) question, and it was marked as correct.
WK taught me 入り口 though. I checked jisho, both are correct. However, WK does not have 入口 as vocab entry.
I was under impression KameSume is “the reverse” of WK. Seems it uses bigger dictionary.
Yes, it’s bigger than WK.
Major Update #7: The un-reversed WaniKani edition!
Phew, this one was a lot of work. Up to now, KameSame has existed as a “Reverse WaniKani” tool, but once I started using it, I really wanted to be able to use KameSame for words that weren’t part of the WaniKani corpus (hence pulling in JMDict’s 180,000 words in Update #6). As a result, however, any words you study in KameSame from JMdict will be a little one-sided—I’m already finding that there are words I can recall in conversation thanks to KameSame but literally can’t read when I see them on signs or in articles due to a lack of recognition practice.
To recap these terms, in KameSame:
- Production: English prompt, Japanese answer
- Recognition: Japanese prompt, English answer
So! With this update, you can now study your ability to recognize Japanese words by providing a valid English meaning. There is a lot to unpack here, so let’s start with the feature and work backwards to setting it up how you need.
Studying recognition skills
When on a recognition card, the study UI does the obvious thing and swaps English for Japanese:
You provide the right English response, and you pass the lesson or review:
That’s about all there is to that. When you complete a set of reviews that encompass multiple skill types, the results will be annotated with
あ badges to indicate whether they were recognition or production skills, accordingly.
Recognition and Production as separate cards
In WaniKani, we practice both meaning recognition and reading production, and the two are a “package deal” for purposes of tracking your knowledge of an item. If you get either wrong, you fail the whole item. The item isn’t complete until you get both right.
KameSame works differently. For the skills you have enabled, each one is learned and then tracked completely separately. It’s possible to have burned recognition on one card while still an apprentice in production, for example. You can see your progress towards each on the item summary page:
Enabling and disabling the feature
By default, recognition will be enabled for all users doing ad hoc & content-based lessons based on your enabled types (e.g. if you have vocabulary enabled but kanji disabled, only vocabulary recognition will be toggled on). However, if you’re sure you never want to practice recognition with KameSame, you can disable it entirely from the accounts settings like this:
Batching by question type
If you don’t have “Batch studies by type” enabled in settings, I strongly urge you to turn it on as a quality of life measure:
When enabled, you clear all of your reviews that require an English keyboard first, allowing you to toggle back to a Japanese keyboard for production just once per session. If you’re using a physical Japanese hardware keyboard or otherwise comfortable toggling rapidly between keyboard modes, maybe experiment with turning this back off for greater surprise/variety during your reviews.
Learning recognition of WaniKani items
Even though the overall feature is enabled by default, you will still only get production lessons via the WaniKani lesson interface unless you opt-in to study both, since you (presumably) are already getting recognition practice for the word using WaniKani.
No such thing as too much practice, though!
There are a few consequences of this feature you may or may not notice:
- So as not to make it too easy to earn a weekly star, each star requires 50 lessons across unique items each week. So if you learn 50 production cards, you’ll get your star. If you learn production & recognition for the same 49 items, you won’t earn your star even though you’ll have completed 98 lessons.
- There are no “reading correct” or “alternate match” answers for recognition cards. Just right and wrong.
- Like WaniKani, you don’t have to type a recognition answer perfectly to be marked correct. The typo distance is relaxed as the meaning gets longer so you’re not having to sweat typing really long meanings too carefully
- Because many JMDict items have long parentheticals in their definitions (e.g. “To run (to a store, etc)”), the scoring mechanism will accept either an exact answer or an answer with all the parentheticals removed. That means if one meaning in JMDict is “To travel (in a vehicle) away from (your own) home” then “to travel away from home” will be an accepted answer
This feature is going to necessitate a lot of follow-on work over the next few months. I imagine we’ll need to:
- Clean up the home/progress page to differentiate different skills across items (right now you’ll see duplicates)
- A synonym (“alternate meaning” to go along with “alternate spelling”) system so that you can type shorter, more memorable meanings—also tie these back to alternate matches of production cards, potentially
- A vocabulary list feature, featuring either or both of curated and custom lists to study from. I’d like to make it really easy to start plugging away at the top 10,000 words, for example. Same goes for studying all the words by JLPT level.