So, still having problems with getting rid of uncommon or not-at-all-used readings for kanji. For example, right now I got a multiple choice question about the reading of this kanji. The right option was “ろう and る“. Problem is, I have deactivated those readings. On the bright sight (this screen is from after the schedule) it at least doesn't seem to mark those readings as known afterwards. But still, they shouldn't be asked when they are deactivated.
Edit: Okay, so it was just a bit slow to update; the on readings were now marked as known. But I don't want to know them! xD
Sorry, just not getting around to my back log. This is definitely not a feature improvement but a bug. Can you please shoot me a few in-quiz reports next time you see these blocked readings coming up in a quiz?
One question about that though. Usually, if I first learn a new kanji usually none of the readings are marked as known (obviously), so if they are already grey, how am I supposed to block them by making the readings unknown? Make them known and then unknown again?
Edit: So basically the problem I have with the current system is, that there really is no difference between a reading that I “blocked" and one I haven't learned yet. I would prefer an explicit “block“ function.
Sorry, one more thing in this context. Within quizzes the readings are all black. First screenshot is inquiz and the one below I took afterwards in the dictionary. Also, I think ぐ is not a reading of this kanji, maybe あた as well isn't.
Edit: This seems to be a bigger issue maybe. The readings marked as known in the tests and in the dictionary seem to vary (regardless of what readings are tested before, so it is not a matter of slow updating or something).
In addition to what I said in the mail; again the question: How can I block readings when I first learn a new term. All readings are grey in that case (for not learned). Do I have to make them all known and then unknown again? It is confusing.
Ok, so let's summarize what is here so we can best decide how to resolve it:
There are three states of readings:
known (in black)
unknown (decided by the site based on rarity of reading)
unknown (user explicitly blocks this reading from appearing)
The two unknowns both appear in grey, so there is ambiguity in what that actually means.
Typically, the site will display all known terms, but if there are only unknown terms because it is a rare kanji to begin with, then it doesn't have way to split them up, and therefore shows all of them.
I think there should be a differentiation visually between 2 and 3:
unknown - decided by the site - grey (same as current)
blocked/unknown - user action - either color the text red, or put a little x mark next to it.
In doing this, the unknown terms could be toggled to blocked terms before they are ever quizzed.
Hey, first of all thank you for taking your time to try to solve this, I really appreciate that:
I agree there should be a differentiation between 2 and 3. Red seems to make sense and would be really clear; a little x in my opinion would cause ambiguity again, since the little x is already used for rare or obsolete readings (afaik). Alternativly maybe a strikethrough would be an option as well since it doesn't highlight them as much as the red colour.
BUT: Here are some things though that I noticed an that are kind of connected to this issue for me: First of all I wanted to ask what the “hide rare kanji readings“ option in the settings does (I didn't see a difference)? Secondly, I noticed obsolete readings marked with x being quizzed also when there are standard readings already known (black). I am not sure but I suspect this might have to do with the different quizzing methods (choose on/kun-reading, drag and drop); so for example, if there is no kun known because they are all x'ed it might just take one anyway in order to be able to quiz you (afterwards reading is marked as known)?. Thirdly, since the site already knows a good deal of obsolete readings (little x's), wouldn't it be possible to add a function that blocks them all (marking them red; could be unblocked again of course), because usually it's those reading one would most likely block anyway.
Okay, that was lots of text but it should be all. Again, thanks a lot!
1. I am testing the ui change in the dictionary only - suppressed readings (marked as unknown) should have a little icon next to them. Let me know what you think.
2. Well, it sounds like the setting isn't working - what you are requesting is what is already happening (or should be, rather) - readings marked with an x will not appear in the quizzes regardless of quiz type. Try shooting me a few examples of this happening (inquiz report) so I can check it out!
Edit: this does not yet work for blocking readings on terms you have not yet studied. This needs to be implemented. Also, it does not yet immediately appear after checking the box to block it - it takes a refresh. This is only because I'm testing it out - it will act appropriately when I've made a final choice on how to implement it.
Last edit - a light red (grey + red) might be a better way to do this without adding more text to the screen.
I think it looks best now (underline), and utilizes the same styling for unknown words in sentences.
Also, I didn't answer part of your other question. If terms are blocked (manually, or because they are rare and the setting is on), as far as the kanji lists and quizzes are concerned, that reading doesn't exist. In other words, there is no kind of logic saying "Hey, I need kunyomi for this quiz, but only blocked ones exist. Better choose one of those." Instead, it'll say "Hey, this term has no kunyomi. Can't use it for this quiz."
Hey, thanks for implementing this as a test! I think the underlining looks good. I noticed that it doesn't work for all readings though. I give one example screenshot (it worked for the two other readings, but not the third one; for other kanji it is the same, some work, some don't, or all don't work or maybe there is only one and it doesn't work; so far I would say it didn't work for about a fifth of the terms I tried). Or is this what you meant with (“this does not yet work for blocking readings on terms you have not yet studied. This needs to be implemented.")? Isn't marking as learned equal to actually learn in a quiz?
Thanks for clarifying how the readings are (or aren't pulled) for quizzing; so it doesn't has anything to do with that.
And, regarding that global "hide readings" setting: I think the last mail you replied to me and said you replied here was about exactly that. The reading mentioned in that report (it was for 挨) was also about a reading that is marked with an x.
Edit: Forgot one thing: So far this only works in the dictionary right? In the quizzes and in the term overview for a schedule the underlining doesn't appear.
Edit2: You can't (easily) undo the block for the underlined readings are not clickable.
And two other things besides this (I reported the first one here in quiz): Just got quizzed on a blocked reading (こうむ.る):
After that I marked the term as correct (cos it asked me something I am not learning) and it produced this strange double status. Blocked and known at the same time. Also, before I did this I blocked the other kun readings and after marking it correct those became unblocked again (grey, not underlined).
The second thing (not reported): Sometimes the status of the readings differ between inquiz/overview and dictionary (also after refreshing):
I've made some changes that I believe will fix most if not all of the issues you mentioned. IF you get another "it is blocked, but it came up in a quiz", please try and send a report before you answer the question. Thanks!
Cool, I'll check it out! And I will try to do the report BEFORE, sadly usually I only notice it afterwards; for I usually think it is me not remembering the reading while actually it is a reading I can't know – well, will have to work on my confidence there I guess. ;)
So at least the "problem" regarding the display of the readings seems to still exist (here the こば):
Also still occuring (just tried): So I got a wrong answer and checked the dictionary in-quiz; I blocked an on and a kun reading, then marked the question as correct (for it was a reading I blocked), result: all readings marked as known afterwards. (of course: easy solution would be to first mark the question correct and then block readings. ;) )