Japanese app confuses multiple definitions

人間 has two pronunciations, じんかん and にんげん, with two meanings. I’m studying one of the two:

When the word came up in 3.3.5 on Android, I got a mixture of the two definitions:

  1. The question itself should be asking me for じんかん since that’s the only one I’m studying.
  2. The definition is for にんげん. (wrong)
  3. The sound played was にんげん. (wrong)
  4. The hiragana is じんかん. (correct)

This is definitely a bug, thanks for letting us know! Have you noticed this on any other words? Some words with the same base writing do have multiple entries with different definitions / readings however it looks like its getting mixed up with where to pull from on this entry.

This is the only word I’ve noticed it with. The sound is what tipped me off, and I frequently study in places where I can’t have the audio turned up.

I take that back. I had a bizarre experience while studying Chinese the other day that I just realized was an instance of this problem.

I am studying both 喂 and 餵. 喂 means “hello”, whereas 餵 means to feed, but in simplified Chinese 喂 takes on both meanings. Skritter’s definitions:

喂: hello (on the phone); hey; to feed
餵: to feed

As far as I am aware these are accurate, showing that the 口 version takes on both meanings, and the 食 version only means feed.

In any case, Skritter asked me to draw “hello (on the phone); hey; to feed”. Seeing “hello”, which is unique to 喂, I started drawing 喂, but got it wrong. It wanted 餵. At the time I thought that was strange, but figured hello was some fringe meaning of 餵 that I wasn’t aware of – Skritter’s definitions can be a bit over-inclusive.

But the reality is I was literally given the definition from one character, and was asked to draw the glyph from another one.

I’m unfortunately not able to reproduce this (even using the specific examples you provided in addition to other cases where this might happen), which makes me think that it might be an issue due to lingering data. I’m hoping this is fixable by clearing your devices app cache, and reinstalling the app.

You’re running an Android device, right? If you aren’t sure how to clear app cache you can refer to: https://www.androidcentral.com/how-and-when-clear-app-cache-or-data-android

Once that’s done, I would delete the app and reinstall it (all of your data is stored on the servers so there’s no risk of losing any progress by deleting the app).

Please let me know how this goes!

Sure, I’ll give it a try.

That said:

  • It’s a brand new phone (less than a week) with a fresh install of both Skritters.
  • Whatever is “lingering” here should never have existed in the first place. In both cases, I’m being presented a mixture of data from two words. All words involved were added weeks, if not months ago.
  • The failure mode is Skritter teaches me things that are wrong, and detecting it depends on me knowing that, which is obviously unreliable. Should I be reinstalling every 2-3 days just in case?
1 Like

Hm. This likely won’t do the trick in that case. I was thinking it might not be such a new device / install. I would go ahead and give this a shot anyway just to get it out of the way!

Here’s another:

My intention is to learn the “female animal” definition of メス:

But the app gives me this:

I remember getting this on my old phone and being puzzled, but didn’t track it down at the time.

We’ve got a task open to dig into your account on the Chinese side. @Jeremy add some notes so we can do the same on the Japanese side, please :slight_smile:

Thanks for all the in-app examples. It’ll help make debugging a lot easier!

1 Like

FYI, the problems persist post-wipe:

1 Like

Learning: 弱 = よわい

The card is visually correct:

The sound played, however, is じゃく.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.